AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 1 Technical Report Summary La Colosa Project An Initial Assessment Report Effective date: 31 December 2021 As required by § 229.601(b)(96) of Regulation S-K as an exhibit to AngloGold Ashanti's Annual Report on Form 20-F pursuant to Subpart 229.1300 of Regulation S-K - Disclosure by Registrants Engaged in Mining Operations (§ 229.1300 through § 229.1305). AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 2 Date and Signatures Page This report is effective as at 31 December 2021. AngloGold Ashanti has recognised that in preparing this report, the Qualified Person may have, when necessary, relied on information and input from others, including AngloGold Ashanti. As such, the table below lists the technical specialists who provided the relevant information and input, as necessary, to the Qualified Person to include in this Technical Report Summary. All information provided by AngloGold Ashanti has been identified in Section 25: Reliance on information provided by the registrant in this report. The registrant confirms it has obtained the written consent of the Qualified Person to the use of the person's name, or any quotation from, or summarisation of, the Technical Report summary in the relevant registration statement or report, and to the filing of the Technical Report Summary as an exhibit to the registration statement or report. The written consent only pertains to the particular section(s) of the Technical Report Summary prepared by the Qualified Person. The written consent has been filed together with the Technical Report Summary exhibit and will be retained for as long as AngloGold Ashanti relies on the Qualified Person’s information and supporting documentation for its current estimates regarding Mineral Resource or Mineral Reserve. MINERAL RESOURCE QUALIFIED PERSON Rudolf Jahoda Sections prepared: 1 – 25 ________________ Responsibility Technical Specialist ESTIMATION Rudolf Jahoda EVALUATION QAQC Rudolf Jahoda EXPLORATION Rudolf Jahoda GEOLOGICAL MODEL Rudolf Jahoda GEOLOGY QAQC Monica Uribe GEOTECHNICAL ENGINEERING Carlos Mejia HYDROGEOLOGY Juan Gomez MINERAL RESOURCE CLASSIFICATION Rudolf Jahoda ENVIRONMENTAL and PERMITTING Ruben Latorre FINANCIAL MODEL Rudolf Jahoda INFRASTRUCTURE Rudolf Jahoda LEGAL Jose Florez METALLURGY Rudolf Jahoda MINE PLANNING Rudolf Jahoda /s/ Rudolf Jahoda AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 3 Consent of the Qualified Person I, Rudolf Jahoda, in connection with the Technical Report Summary for “La Colosa Project, An Initial Assessment Report” dated 31 December 2021 (the “Technical Report Summary”) as required by Item 601(b)(96) of Regulation S-K and filed as an exhibit to AngloGold Ashanti Limited’s (“AngloGold Ashanti”) annual report on Form 20-F for the year ended 31 December 2021 and any amendments or supplements and/or exhibits thereto (collectively, the “Form 20-F”) pursuant to Subpart 1300 of Regulation S-K promulgated by the U.S. Securities and Exchange Commission (“1300 Regulation S-K”), consent to: • the public filing and use of the Technical Report Summary as an exhibit to the Form 20-F; • the use of and reference to my name, including my status as an expert or “Qualified Person” (as defined in 1300 Regulation S-K) in connection with the Form 20-F and Technical Report Summary; • any extracts from, or summary of, the Technical Report Summary in the Form 20-F and the use of any information derived, summarised, quoted or referenced from the Technical Report Summary, or portions thereof, that is included or incorporated by reference into the Form 20-F; and • the incorporation by reference of the above items as included in the Form 20-F into AngloGold Ashanti’s registration statements on Form F-3 (Registration No. 333-230651) and on Form S-8 (Registration No. 333-113789) (and any amendments or supplements thereto). I am responsible for authoring, and this consent pertains to, the Technical Report Summary. I certify that I have read the Form 20-F and that it fairly and accurately represents the information in the Technical Report Summary for which I am responsible. Date: 30 March 2022 Rudolf Jahoda /s/ Rudolf Jahoda AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 4 Contents 1 Executive Summary ................................................................................................................................... 7 1.1 Property description including mineral rights .................................................................................... 7 1.2 Ownership ....................................................................................................................................... 8 1.3 Geology and mineralisation.............................................................................................................. 8 1.4 Status of exploration, development and operations .......................................................................... 8 1.5 Mining methods ............................................................................................................................... 9 1.6 Mineral processing ........................................................................................................................... 9 1.7 Mineral Resource and Mineral Reserve estimates ........................................................................... 9 1.8 Summary capital and operating cost estimates .............................................................................. 10 1.9 Permitting requirements ................................................................................................................. 10 1.10 Conclusions and recommendations ............................................................................................. 10 2 Introduction .............................................................................................................................................. 11 2.1 Disclose registrant ......................................................................................................................... 11 2.2 Terms of reference and purpose for which this Technical Report Summary was prepared ............ 11 2.3 Sources of information and data contained in the report / used in its preparation ........................... 12 2.4 Qualified Person(s) site inspections ............................................................................................... 12 2.5 Purpose of this report ..................................................................................................................... 12 3 Property description ................................................................................................................................. 12 3.1 Location of the property ................................................................................................................. 12 3.2 Area of the property ....................................................................................................................... 13 3.3 Legal aspects (including environmental liabilities) and permitting .................................................. 15 3.4 Agreements, royalties and liabilities ............................................................................................... 16 4 Accessibility, climate, local resources, infrastructure and physiography ................................................... 17 4.1 Property description ....................................................................................................................... 17 5 History ..................................................................................................................................................... 17 6 Geological setting, mineralisation and deposit ......................................................................................... 19 6.1 Geological setting .......................................................................................................................... 19 6.2 Geological model and data density ................................................................................................ 24 6.3 Mineralisation................................................................................................................................. 27 7 Exploration .............................................................................................................................................. 30 7.1 Nature and extent of relevant exploration work .............................................................................. 30 7.2 Drilling techniques and spacing ..................................................................................................... 32 7.3 Results .......................................................................................................................................... 33 7.4 Locations of drill holes and other samples ..................................................................................... 34 7.5 Hydrogeology ................................................................................................................................ 35 7.6 Geotechnical testing and analysis .................................................................................................. 37 8 Sample preparation, analysis and security ............................................................................................... 39 8.1 Sample preparation ....................................................................................................................... 39 8.2 Assay method and laboratory ........................................................................................................ 42 8.3 Sampling governance .................................................................................................................... 42 8.4 Quality Control and Quality Assurance .......................................................................................... 45

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 5 8.5 Qualified Person's opinion on adequacy ........................................................................................ 46 9 Data verification ....................................................................................................................................... 46 9.1 Data verification procedures .......................................................................................................... 46 9.2 Limitations on, or failure to conduct verification .............................................................................. 47 9.3 Qualified Person's opinion on data adequacy ................................................................................ 47 10 Mineral processing and metallurgical testing ......................................................................................... 47 10.1 Mineral processing / metallurgical testing ..................................................................................... 47 10.2 Laboratory and results ................................................................................................................. 47 10.3 Qualified Person's opinion on data adequacy .............................................................................. 49 11 Mineral Resource estimates .................................................................................................................. 49 11.1 Reasonable basis for establishing the prospects of economic extraction for Mineral Resource ... 49 11.2 Key assumptions, parameters and methods used ........................................................................ 51 11.3 Mineral Resource generation steps .............................................................................................. 55 11.4 Mineral Resource classification and uncertainty ........................................................................... 60 11.5 Mineral Resource summary ......................................................................................................... 64 11.6 Qualified Person's opinion ........................................................................................................... 66 12 Mineral Reserve estimates .................................................................................................................... 66 13 Mining Methods ..................................................................................................................................... 66 14 Processing and recovery methods ......................................................................................................... 66 15 Infrastructure ......................................................................................................................................... 66 16 Market studies ....................................................................................................................................... 67 17 Environmental studies, permitting plans, negotiations, or agreements with local individuals or groups .. 68 17.1 Permitting .................................................................................................................................... 68 17.2 Requirements and plans for waste tailings disposal, site monitoring and water management ...... 69 17.3 Socio-economic impacts .............................................................................................................. 69 17.4 Mine closure and reclamation ...................................................................................................... 69 17.5 Qualified Person's opinion on adequacy of current plans ............................................................. 69 17.6 Commitments to ensure local procurement and hiring ................................................................. 70 18 Capital and operating costs ................................................................................................................... 70 18.1 Capital and operating costs .......................................................................................................... 70 18.2 Risk assessment .......................................................................................................................... 70 19 Economic analysis ................................................................................................................................. 70 19.1 Key assumptions, parameters and methods ................................................................................ 70 19.2 Results of economic analysis ....................................................................................................... 70 19.3 Sensitivity analysis ....................................................................................................................... 70 20 Adjacent properties ................................................................................................................................ 70 21 Other relevant data and information ....................................................................................................... 71 21.1 Inclusive Mineral Resource .......................................................................................................... 71 21.2 Inclusive Mineral Resource by-products ....................................................................................... 71 21.3 Mineral Reserve by-products ....................................................................................................... 71 21.4 Inferred Mineral Resource in annual Mineral Reserve design ...................................................... 71 21.5 Additional relevant information ..................................................................................................... 71 21.6 Certificate of Qualified Person(s) ................................................................................................. 71 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 6 22 Interpretation and conclusions ............................................................................................................... 72 23 Recommendations ................................................................................................................................. 72 24 References ............................................................................................................................................ 73 24.1 References .................................................................................................................................. 73 24.2 Mining terms ................................................................................................................................ 76 25 Reliance on information provided by the Registrant ............................................................................... 79 List of Figures La Colosa project location and project property map. ................................................................................. 13 La Colosa integrated property map including surface rights and environmentally sensitive areas. .............. 14 EIG-163 concession contract. ..................................................................................................................... 15 La Colosa regional geology and geochronological data. ............................................................................. 20 La Colosa local geology .............................................................................................................................. 21 La Colosa deposit geology ......................................................................................................................... 21 La Colosa rock petrography ........................................................................................................................ 23 La Colosa 3D lithology and WE section. ..................................................................................................... 27 La Colosa grade shells. .............................................................................................................................. 29 Sulphide distribution based on mapping and detailed logging. .................................................................... 29 Plan view with drill hole locations. ............................................................................................................... 34 Drill hole and exploratory data analysis. ..................................................................................................... 35 Location of hydraulic tests, piezometers, flow stations, rain gauges, and weather stations ......................... 36 Precipitation at La Colosa ........................................................................................................................... 36 Geotechnical drill holes across La Colosa .................................................................................................. 38 WOL and FFG flowsheets .......................................................................................................................... 48 Geometallurgy comminution - tonnes per hour (tph) geometallurgical throughput estimates. ..................... 54 La Colosa inclusive Mineral Resource grade and tonnage curve ................................................................ 55 Estimation and modelling techniques. ......................................................................................................... 56 Colosa estimation domains ......................................................................................................................... 57 Preliminary matrix for ARD indicators and criteria for geochemical domain definition ................................. 60 Mineral Resource classification .................................................................................................................. 62 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 7 1 Executive Summary 1.1 Property description including mineral rights The La Colosa prospect occurs within a forest reserve declared by the Colombian Government in 1959. Greenfields exploration activities commenced in 2006 and led to an exploration drilling program starting in March 2007. The existence of a multi-million-ounce gold deposit was evident in mid-2007 and drilling activities were increased. The local environmental authority ordered the suspension of exploration activities on 26th February 2008 and asked for the completion of a temporary forest subtraction process. Drilling restarted on 16th August 2010, using temporary forest extraction platform sites for drilling. Additional platform permits were also issued for open forest sites for metallurgical and Mineral Resource drilling. Using these sites the Pre-Feasibility Study (PFS) drilling was completed in December 2014. Geotechnical-hydrogeological- condemnation drilling was carried out from 2015 to 2017 at the Site 11 infrastructure area. An additional 800m Mineral Resource compliance drilling program was conducted in early 2017. In April 2017, AngloGold Ashanti Colombia decided to put the project on care and maintenance. The Qualified Person (QP) acted as Geology Manager till 30th July 2017 compiling geological information into a PFS ‘Light Report’. The Mineral Resource block model was updated with the final wireframe models, the drilling, and the geometallurgy information. The project has been in force majeure since October 2017, and from that date AngloGold Ashanti Limited’s legal department has taken custodianship of the property. From April 2017 neither exploration nor environmental monitoring has been carried out. Drill platforms, piezometers, weather stations and the El Aceituno core shed have not been visited since that date. Secondary forest growth has been substantial. A desktop study for a small-scale mining project was conducted in late 2017, a temporary forest subtraction request was submitted in 2018. A second desktop study was carried out in 2020-2021 leading to an assessment memorandum for alternative mining scenarios, and this included an option for filtered tailings design. AngloGold Ashanti has requested the renewal of force majeure till June 2022 and this has been granted. The data presented in this report dates back to the exploration activities carried out between 2015 - 2017, with some discussion of concepts presented in the June 2021 assessment memorandum. The project is located 150km west of the Colombian capital city, Bogota, and 30km west of the major town of Ibague, which is the capital of the Tolima Department. Ibague is the location of local government entities monitoring the project. Cajamarca, a small rural town of 12,000 inhabitants is located a 14 km drive east of the project, and at an elevation of 1800m. The La Colosa camp, on the gentler eastern slope of the La Colosa ridge, was constructed at an elevation of 2800m. The La Colosa ridge at its highest elevation reaches 3450m. Vegetation at the project site is partially high Andean forest and grassland (originally grown for cattle grazing). The high Andean forest gives way to more scrubby vegetation at an elevation of around 3400m. The project collected climate data from three weather stations: San Antonio (located on the La Colosa ridge), Mirador on the eastern slope of the ridge, and Manantial to the west at the El Diamante slopes. Cumulative annual rainfall varies between 940 and 1,300mm for Manantial, between 1,100mm and 1,520mm for Mirador, with the total cumulative rainfall at San Antonio in 2015 being 85mm. The initial PFS studies commenced in September 2008. In 2014, the project team presented an update of the initial conceptual study and reframed the PFS deliverables. Mining activities (exploration, construction, exploitation, and closure) are governed by Colombian mining law. To obtain a mining right, a concession contract must be signed by the mining authority (National Mining Agency). Concession contracts are granted for 30-year periods, extendable by another 30 years, and with a right of first refusal of an additional 30 years once the second extension ends. The La Colosa project is covered by the EIG-163 mineral title (tenement) with an area of 9,210.11 hectares. Initially, there were 6 adjacent concession contracts and these were integrated into the EIG-163 concession contract in March 2017. AngloGold Ashanti Colombia signed the EIG-163 mining contract for an initial term of 20-year with the National Mining Agency. The 10 years of exploration carried out before the integration were deducted from the maximum possible of 30 years. The new EIG-163 started again in "Year 1" of exploration. The EIG-163 contract is fully located in a forest reserve zone aimed to promote the timber industry. To develop mining activities in the forest reserve zone, a forest reserve subtraction permit must be issued by the Environmental ministry. Due to the delay in issuing the forest reserve subtraction permit for the La Colosa AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 8 project, force majeure conditions were recognised and declared by the National Mining Agency in June 2017. Concession contract obligations are now suspended and not enforceable during force majeure, and the time while in force majeure does not count against the permit license, therefore, the exploration tenement continues to be on its first year of exploration. Force majeure is in force until June 2022 and can be renewed if the lack of environmental permits continues. Exploration activities have not been carried out since April 2017 and operational activities are suspended. Only care and maintenance activities are carried out. The timeframe for exploration at La Colosa is 11 years (minus 1 month before force majeure was declared) before entering the construction phase. AngloGold Ashanti Colombia is obliged to report every semester on advances of the force majeure condition, and audits by the National Mining Agency. Similarly, the regional environmental authority, Cortolima, as the local environmental authority, monitors suspension of activities at the project. The extension of force majeure needs to be requested and justified on an annual basis. 1.2 Ownership AngloGold Ashanti Colombia is the only owner of the EIG-163 exploration/mining title. In 2017, the company signed the integrated exploration contract for EIG-163, which contains the La Colosa deposit. The contract marks the beginning of 1st year of exploration. The contract is currently in force majeure with the clock stopped due to the lack of environmental permits (e.g., subtraction permit) to conduct exploration activities. The operator of the project is AngloGold Ashanti Colombia. 1.3 Geology and mineralisation The deposit forms part of what is generally known as the Middle Cauca Metallogenic Belt. The best-known porphyry (Cu-Au, Au-Cu, Au) and intermediate sulphidation Au-Ag deposits in the Middle Cauca Metallogenic Belt are the Marmato and the Buritica mining operations. Advanced exploration studies exist for the Quebradona, the Titiribi, and the La Mina deposits. Exploration highlights with long intersections greater than 100m at grades greater than 1g/t are being reported for Tesorito close to Quinchia (explored by Los Cerros). The La Colosa porphyry complex consists of three intrusive stages: The early, the intermineral, and the late- stage magmatic event. The U-Pb ages obtained range between 7.4Ma and 8.5Ma indicating the emplacement of the early, intermineral and late intrusive porphyry stocks occurred during a very short time span of about 1.1 million years. The emplacement of the La Colosa and San Antonio porphyry stocks caused contact metamorphism that transformed the proximal country rocks into hornfels. Recent volcanism in the Central Cordillera accounts for an ash cover varying between a few and 15m thick. The source is the Cerro Machin stratovolcano, located about 17 km to the east of the La Colosa project site. The La Colosa porphyry Au deposit has nine defined broad hydrothermal alteration assemblages: Sodic- calcic alteration, potassic alteration, quartz-sericite alteration, sericitic alteration, chloritic alteration, propylitic alteration, intermediate argillic alteration, silicification, and supergene argillic alteration. Eight types of porphyry veinlets have been recognised: early biotite veinlets, A-type veinlets, B-type veinlets, M-type veinlets, N-type veinlets, AC-type veinlets, S-type veinlets, and D-type veinlets The veinlets occur in the early, intermineral and late porphyries, as well as in the schistose wall rock. In addition, there are veinlets representing a younger, late or post-porphyry event. At the La Colosa project, three gold mineralising events have been observed. The first gold event (porphyry event) is associated with the magmatic pulses of the intrusive complex. The early intrusive porphyries and early intrusive breccias show gold grades ranging from 0.75 to 1.0g/t. A-type and S-type veinlets are abundant and potassic alteration is the common alteration type. The second gold mineralisation event is associated with the N-trending extensional faults that crosscut the early, intermineral, and late stage porphyries and the wall rock (younger event). The third event, seen locally, is related to supergene argillic- FeOx alteration of pyrite-rich/pyrrhotite-poor associated with the late porphyry. 1.4 Status of exploration, development and operations The last integrated Steering Committee meeting took place in October 2016 and since March 2017 neither additional drilling nor exploration activities have been carried out. Much of the information discussed in this

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 9 report has been taken from the author´s 2017 Mineral Resource update and the PFS Light geology report. Technical reports for infrastructure designs and hydrogeology were prepared by Golder Associates, plant designs by HATCH and most metallurgical studies have been carried out by SGS Lakefield. The environmental studies consist of acid rock drainage (ARD) data (static and kinetic for potential waste material), climate studies (five meteorological stations in the project area), rain gauges, and river flow measurements. 1.5 Mining methods During past PFS’ of the project, both open pit and underground mining scenarios have been considered. Open pit mining (with potentially minor underground mining) is the preferred mining method. Initial sensitivity studies for annual throughputs ranging from 6 Mt per year to 26 Mt per year have been carried out. Geotechnical studies for pit designs are at advanced PFS level and pit hydrogeology is at an initial PFS level according to the company standards. The earlier mining studies have used pit optimisations for different gold prices, however, did not advance to more detailed open pit designs. 1.6 Mineral processing The mineralisation at La Colosa consists of two principal ore types, diorite, and schist. Both are sulphidic and there is a minor amount of oxide material. Comminution test work conducted in 2011 and 2012 showed the ore to be relatively hard and suitable for milling by a SAG/ball mill circuit. Recovery test work included mineralogy, diagnostic leaching, gravity separation, cyanidation, cyanide detoxification, and flotation. Comminution and recovery variability test work was conducted on a large number of samples in 2013 and 2015. In 2015 further test work was conducted to examine the viability of a circuit with flotation. This included fine milling of the flotation concentrate and cyanidation test work on flotation tailings and the milled flotation concentrates. The test work showed that the ore is amenable for treatment by either a conventional whole ore leach (WOL) treatment circuit or a circuit involving flotation and cyanidation of the flotation concentrate after fine milling as well as cyanidation of the flotation tailings (flotation fine grind, or FFG). These flowsheets are depicted in the figure showing the WOL and FFG flowsheets in the main metallurgy chapter. The base case flowsheet is WOL at a primary P80 grind size of 75µm and a gold recovery of 83.7%. The alternate case is FFG at a primary P80 grind size of 105µm and a gold recovery of 84.1%. The base case throughput is 23Mtpa achieved with one SAG/ball milling line. Mill sizing was undertaken by Hatch and is in line with mill supplier calculations. The alternate FFG circuit can achieve a throughput of 26Mtpa with the same milling circuit if the primary grind size is coarser. Plant layouts were undertaken by Hatch at the Los Pinos site and the El Diamante site and indicated that the plants can be constructed on these sites but will incur significant earthwork costs. Smaller throughput cases of 15, 10, and 6 Mtpa were evaluated in a desktop study using conventional factoring methods. For tailings disposal, filtration is considered. Although no filtration test work has been undertaken to date, the size distributions of the plant tailings (WOL tailings or flotation tailings and milled concentrate tailings) are within the typical suitable range for large scale industrial filtration equipment. In the case of the FFG circuit, the concentrate is milled to a P80 of 25µm which is common for a reground concentrate to be filtered in large industrial mineral processing circuits. The project is currently at an early stage however flotation of sulphide ore is being considered as a treatment option. 1.7 Mineral Resource and Mineral Reserve estimates La Colosa does not report Mineral Reserve, therefore, exclusive and inclusive Mineral Resource are the same. Exclusive Mineral Resource (Base: 2017) La Colosa Tonnes Grade Contained gold as at 31 December 2021 Category million g/t tonnes Moz Measured - - - - Indicated 833.49 0.87 726.31 23.35 Measured & Indicated 833.49 0.87 726.31 23.35 Inferred 217.89 0.71 154.86 4.98 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 10 1.8 Summary capital and operating cost estimates La Colosa does not report capital and operation estimates. 1.9 Permitting requirements The EIG-163 contract is fully located in a forest reserve zone aimed to promote the timber industry. In order to develop mining activities within the forest reserve zone, a forest reserve subtraction permit must be issued by the Environmental ministry. A forest reserve subtraction permit application was filed by AngloGold Ashanti in 2018 to advance its studies through both PFS and Feasibility Studies (FS). The pending subtraction permit is the principal permit required to advance the La Colosa study for drilling platforms and their access. In the past, the Environmental ministry permitted platforms as specific locations with a certain flexibility to adjust for topography and access. Feasibility studies will require area permits, such as broadly defined infrastructure locations to allow for geotechnical and hydrogeological drilling as defined by engineering contractors. Area permits for geotechnical and hydrogeological drilling are also required in the areas of influence for environmental permitting. Additional permits to be considered are for water concessions, water discharge, occupation of water channels (creeks), and, if necessary, temporary forest clearing and the use of timber. The permits required for mine construction and mine operation are an Environmental License (EIA), an approved Construction and Mining Plan (PTO), and permission for definite forest clearing. The project will continue as an exploration project in Year 1 once force majeure ends. The company will then submit a report to the National Mining Agency detailing all surface and subsurface exploration it plans to carry out. That report includes schedules and investments. There is an environmental mining guide that exploration activities must adhere to and accordingly a general environmental impact estimate will be carried out for surface and sub-surface exploration. The environmental management plan will be prepared and included in the platform permitting process. While all permits can be achieved, the required timeline can only be estimated with a high degree of uncertainty. 1.10 Conclusions and recommendations Based on the current level of exploration, the La Colosa deposit contains a combined Indicated and Inferred Mineral Resource of 1,051.4 Mt at a grade of 0.84g/t containing 28.3 Moz Au. The La Colosa deposit remains open to the north and at depth. Small upside potential exists around San Antonio and for greater than 1g/t surface samples east of the La Colosa creek. Mineral Resource drilling essentially ceased in 2014 and, consequently, for similar gold price and pit optimisation parameters, there is little change in the estimate of Mineral Resource compared to previous years. In general, geochemical analyses of La Colosa ore have low arsenic (average of approximately 35 ppm), low Cu (average less than 400 ppm) and low in Mo (average less than 45 ppm). Average sulphur concentration for grade shells varies between 2 and greater than 4% making ore and waste potentially acid generating and metal leaching. The site geology is controlled by steeply east-dipping porphyry intrusions, schistose hornfels and a structural geological framework. Post-mineral, recent volcanic ash overlay a paleo-topography. Geotechnical domains are massive, blocky, and fractured intrusives, similarly, massive, blocky, and fractured metamorphic rocks. Additionally, the weathered material and the volcanic ash have been modelled, both are subject to natural landslides. The La Colosa orebody is amenable to bulk, open pit mining methods. Open pit mining is assumed to be conducted using conventional truck and shovel methods. The mineral processing plant, tailings and waste rock facilities are planned to be located on the mountain. Pit optimisation was carried out assuming ore throughput of 23Mt/year, whole ore leach with recoveries of 82% for porphyry, 83% for schist and 87% for oxide. The applicable Mineral Resource gold price at 2017 was $1,400/oz, with the discount rate at 6.6%. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 11 Risks: The delineation of the Los Nevados Paramo by Resolution 1987 in November 2016 puts close to 14Moz of the Mineral Resource at risk. It is recommended that AngloGold Ashanti resumes detailed studies of ecosystem processes and services. The failure to grant environmental permits for site operation held up site operations and force majeure was accepted by the government. Force majeure has been extended till June 2022. Mining: At over 800m high, the north wall of the ultimate pit will be a very significant slope. Insufficiently studied rock structures and/or adverse groundwater conditions could potentially impact slope angles and reduce the contained Mineral Resource and project life. Plant layout work indicated that a 23Mtpa whole ore leach circuit can be constructed at the El Diamante site, however this will require a very high level of earthworks. Additional plant layout work is required to optimise the layout with respect to topography. Initial designs indicate that the slope cuts up to 100m high into potentially acid-generating metamorphic rock. Over the past decades, Colombia has had a reputation in the public markets for instability and changing public policies, and long permitting timelines, in particular for large scale mining projects. Opportunities: Geometallurgical studies indicate comminution parameters can be built into the Mineral Resource block model, and, thereby, contribute to the optimisation of pit shells and mining sequences. Geological studies have advanced to potentially include localized uniform conditioning in the future Mineral Resource model. This allows more selective mining optimisations at variable SMU sizes. Mining: Grade streaming and stockpiling may improve the economics of studied scales of operations. Metallurgy: The La Colosa ore responds well to flotation. The flotation-fine-grind study can achieve higher gold recoveries, and from the environmental perspective, concentrates sulphides in a smaller mass which has advantages in controlling potential acid generation from tailings. Variability sampling should proceed. Recommendations: The QP recommends updating the Mineral Resource estimate by including the localised uniform conditioning approach and the geometallurgical proxies. Pit optimisation should include an update for geotechnical sectors and pit slope angles. The El Aceituno core shed has been shut down for almost five years. Corrective maintenance for core boxes is required. At La Colosa site, the excessive growth of secondary vegetation must be controlled by sustainable measures. 2 Introduction 2.1 Disclose registrant The registrant for the report is AngloGold Ashanti Limited. 2.2 Terms of reference and purpose for which this Technical Report Summary was prepared The purpose of the report is a maiden Mineral Resource Technical Summary report based on S-K 1300 for the first NYSE Mineral Resource reporting. The Mineral Resource is reported as at 31 December 2021 and it is reported in situ, contained within the Mineral Resource optimised shell. This report is an Initial assessment report. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 12 Terms of reference are following AngloGold Ashanti Guidelines for the Reporting of Exploration Results, Mineral Resource and Ore Reserve and based on public reporting requirements as per regulation S-K 1300. Although the term Mineral Reserve is used throughout S-K 1300 and this document, it is recognised that the term Ore Reserve is synonymous with Mineral Reserve. AngloGold Ashanti uses Ore Reserve in its internal reporting. The Technical Report Summary aims to reduce complexity and therefore does not include large amounts of technical or other project data, either in the report or as appendices to the report, as stipulated in Subpart 229.1300 and 1301, Disclosure by Registrants Engaged in Mining Operations and 229.601 (Item 601) Exhibits, and General Instructions. The qualified person must draft the summary to conform, to the extent practicable, with the plain English principles set forth in § 230.421 of this chapter. Should more detail be required they will be furnished on request. The following should be noted in respect of the Technical Report Summary: • All figures are expressed on an attributable basis unless otherwise indicated • Unless otherwise stated, $ or dollar refers to United States dollars • COP refers to Colombian peso • Group and company are used interchangeably • Mine, operation, business unit and property are used interchangeably • Rounding off of numbers may result in computational discrepancies • To reflect that figures don’t represent precise calculations and that there is uncertainty in their estimation, AngloGold Ashanti reports tonnage, content for gold to two decimals and copper, content with no decimals • Metric tonnes (t) are used throughout this report and all ounces are Troy ounces • Abbreviations used in this report: gold – Au • The reference co-ordinate systems used for the location of properties as well as infrastructure and licences maps / plans are latitude longitude geographic co-ordinates in various formats, or relevant Universal Transverse Mercator (UTM) projection. 2.3 Sources of information and data contained in the report / used in its preparation The source of information and data contained/used in the report are the project database and the internal AngloGold Ashanti PFS and Conceptual study reports. The data is representative of exploration carried out since the onset of exploration activities. The structural geological model of the project has been constructed by iC Consulenten. Core scanning was contracted to Geospectral Imaging. Infrastructure and metallurgical study reports were prepared by Golder Associates, Hatch, and SGS Lakefield. 2.4 Qualified Person(s) site inspections La Colosa had a stable project team until the change to care and maintenance in 2017. Neither additional exploration nor audits have been carried out since then. The geological, geotechnical, and hydrogeological drilling was supervised by the Geology Manager since start-up. Metallurgical and infrastructure studies have been contracted to recognised international consultants. External audits for Mineral Resource estimation and Mineral Resource classification have been performed by QG Australia and by Bloy Mineral Resource Evaluation. The IGTRP (Independent Geotechnical Tailings Review Panel, A Robertson, G Fernandez, L Smith) conducted evaluation meetings and prepared reports for infrastructure designs. 2.5 Purpose of this report This is the maiden reporting of the Technical Report Summary (according to S-K 1300) for this project. There are no previously filed Technical Report Summaries for this project. This Technical Report Summary supports the declaration of Mineral Resource for AngloGold Ashanti´s La Colosa project. 3 Property description 3.1 Location of the property The La Colosa gold project is located in the Central Cordillera of Colombia, approximately 8 km WNW of the town of Cajamarca and 30 km west of the city of Ibague, capital of the Department of Tolima. The prospect occurs in the proximity of the highway connecting Bogota (150 km east of the prospect) with the port of

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 13 Buenaventura near the city of Cali. The project employs the UTM coordinate system following corporate standards. The government reporting continues in the Magna Sirgas coordinate system defined by the Instituto Geografico Agustin Codazzi (IGAC). La Colosa project location and project property map. In Colombia, mining activities are regulated by the Mining Code (Law 685 of 2001), with mining activities being part of the policies set forth by the National Development Plan of each local administration. The Colombian state is the owner of the subsoil and the non-renewable natural resources of "any kind and location, lying on the soil or subsoil, in any natural physical state", and such property is "inalienable and imprescriptible." The state grants the right to exploit non-renewable resources by means of concession contracts, which materialise in the form of mining titles. The construction and assembly, operation, closure, and abandonment phases require approval of the environmental license. The license sets forth the requirements and conditions that must be met by private individuals regarding the use and management of natural resources and establishes the different obligations that must be complied with in order to prevent, mitigate, correct, compensate and manage the impact produced by the authorised work or activity. The main risks at the La Colosa project are related to the Los Nevados paramo delineation, the timely approval of permits required to execute exploration activities (current delays have resulted in the declaration of force majeure, as well as the possibility to achieve the Social License to Operate (SLO) and the socio- political enablement of the project. 3.2 Area of the property The La Colosa gold project is covered by the exploration title EIG-163 comprising 9,210.11 hectares. The original title was registered in the National Mining Register on 1st March 2007 during AngloGold Ashanti’s greenfields exploration program. In 2017, after 10 years of exploration, six adjacent exploration titles were integrated and formed the base for the new EIG-163 contract. The new EIG-163 contract was registered on 19th May 2017 with this resetting the timeline to "Year 1". The initial 10 years of exploration has been from AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 14 the maximum 30 years exploration/mining contract. Time does not advance during force majeure; therefore, the exploration title remains in "Year 1". For La Colosa, AngloGold Ashanti has signed a 20-year exploration/mining contract with the National Mining Agency (ANM) of Colombia and mining can be extended for additional 30 years. It is assumed the exploration permit provides sufficient area for open pit design and can cater for plant and infrastructure construction. The Central Cordillera forest reserve, declared by the Colombian Government in 1959, covers the entire EIG-163 exploration permit. Also, the EIG-163 exploration permit is partially overlain by two formally delineated paramo ecosystems, the Los Nevados in the north and the Chili-Barragan paramo to the southwest. AngloGold Ashanti Colombia has purchased 37 farms totalling 6,143 hectares. Some 3,329 hectares of land is on loan to local farmers for cattle or agricultural use. Some 2,020 hectares of these surface rights are located inside the delineated Chili-Barragan paramo, and 731 hectares inside the delineated Los Nevados paramo. Carton de Colombia (a SMURFIT KAPPA, an international paper producer company) maintains pine forests in the area. The Carton de Colombia land has been studied for possible infrastructure designs. A buffer zone around La Linea Highway tunnel defines an environmentally sensitive area to the EIG-163 exploration title. La Colosa integrated property map including surface rights and environmentally sensitive areas. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 15 3.3 Legal aspects (including environmental liabilities) and permitting Mining Rights: AngloGold Ashanti Colombia registered the integrated EIG-163 concession contract on 19th May 2017. The mining concession contract grants rights on subsoil for prospecting and exploration. If the environmental license is issued by the National License Authority and the construction and assembly permit is granted by the Mining ministry, the concession contract can move to exploitation phase. The figure below summarises the activities, decision points, and stage gates of the concession contract. EIG-163 concession contract. Key dates are: • Start date: March, 2017- Initial Term: 20 years (expiring March, 2037) • Extension (if required) - 30 years (to March 2067). • Right of first refusal: 30 years (March 2097). EIG-163 concession contract is fully owned by AngloGold Ashanti and no partnership or joint venture is involved. Land access: Soil use by the mining company can be negotiated with landowners or, in the case of lack of interest to negotiate, mining law authorises land expropriation. Land access can be achieved through land purchase or with mining easements. The mining easements are to allow the mining industry´s efficient exercise of mining- related activities in all its phases and stages. It may be possible to impose all necessary easements on properties located inside or outside the mining title area. The easements can be negotiated directly with the landowners or legally imposed through administrative and judicial litigation. La Colosa project has acquired almost 80% of lands required for previous project designs. If the project is redefined, the land acquisition must be reviewed. EIG-163 is located in the forest reserve zone (allocated for timber industry development). Mining activities require the approval of a forest reserve subtraction permit. In the past, the La Colosa project study team has carried out exploration under two temporary forest reserve subtraction permits granted by the Environmental ministry (MADS): 6.39 hectares in May 2009 (Resolution 814 including two amendments), 36 hectares in 2014 (Resolution 1731). Both permits specify systematic monitoring and reporting requirements. In 2017, the Environmental ministry denied a temporary forest extraction of 169 hectares requested for EIA studies. The study requirements mostly related to infrastructure, geotechnical, and hydrogeological tasks. In March 2018, AngloGold Ashanti subsequently filed a new forest subtraction permit request. The application is still active, and no decision has been taken by the Environmental ministry. In 2021, using knowledge gained from the Quebradona project it is seen that point permitting (platforms) will need to change to area permitting to allow AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 16 all future FS/EIA drilling. The actual permitting strategy needs to include a timeline covering exploration activities up until the EIA submittal. EIG-163 relates to an integrated concession contract signed by the National Mining Agency and AngloGold Ashanti Colombia SAS. The mineral tenement has been registered at the mining cadastre, hence mining rights are fully secured. To develop mining activities in the forest reserve zone, a subtraction permit is mandatory. A forest reserve subtraction permit request was filed in March 2018 and approval is still pending. The absence of this permit prevents mining activities until granted by the environmental ministry. The lack of this permit resulted in a force majeure condition that was formally declared and recognised until a subtraction permit is granted. Six- monthly reports are filed with the National Mining Agency declaring neither exploration nor monitoring is being conducted during force majeure. Paramo Delimitation: In November 2016, the Colombian government issued Resolution 1987 delineating certain wetlands or moorlands as environmentally important protected areas. The resolution includes certain areas in and around the La Colosa project tenement area. In alpine moorland or paramo areas, mining activities are prohibited. This could potentially adversely impact the design, operation, and production of the mining project at the La Colosa project. On 12 June 2017, AngloGold Ashanti Colombia filed a judicial claim against the Environmental Colombian Ministry in the Administrative Court of Cundinamarca to annul Resolution 1987 on technical and legal grounds. The lawsuit was admitted on 30 April 2019 and responded to by the Ministry of the Environment on 16 August 2019. On 2 September the claim was amended to incorporate an expert report to support, from a technical standpoint, that the Paramo Los Nevados delimitation was flawed. There have been no further developments in the case since the end of 2019. Local permits such as water concessions, water disposal, riverbed use, forest and timber use, among others, must be requested from the environmental local authority. An eventual mining scenario requires the approval of an environmental impact assessment (EIA) study and the approval of a mining plan (PTO), the expedition of water permits, and a social license to operate. Additional permits are needed if construction or other engineering activities are required outside the tenement area. Also, invasive studies and constructions within the buffer zone of the La Linea tunnel or in proximity to the national highway are conditional on the authorisation of INVIAS (Instituto Nacional de Vias, the National Road entity). Additional permits for exploitation are required for power lines built from a central transformer station, and the use of explosives requires authorisation from the National Defence Ministry. 3.4 Agreements, royalties and liabilities Royalty: Colombian Law establishes that any exploitation of non-renewable natural resources owned by the state generates a royalty as a mandatory consideration. This consists of a percentage, fixed or progressive, of the exploited gross product of the mining title and its by-products, calculated or measured at the head or edge of the mine, payable in cash or in kind. The current percentage for gold and silver is 4% based on 80% of the international price, which equates to a payable rate of 3.2%. Furthermore, it is calculated as a 4% Royalty Rate on 80% of international London Metal Exchange (LME) price according to Unidad de Planeación Minero Energética (UPME) official reports on total gold and silver production. UPME is a special administrative unit at a national level, it is of a technical nature, and it is ascribed to the Ministry of Mines and Energy. It carries out sustainable planning and development for the mines and energy sectors in Colombia, formulating state policies, and performing information processing and analysis. The General System of Royalties (Article 15 of Law 1530 of 2012) orders the National Mining Agency (ANM) to dictate the terms and conditions for the determination of the base prices for the settlement of royalties and compensation resulting from the exploitation of non-renewable natural resources. According to law, any person or entity subject to the payment of mining royalties on the exploitation of minerals is obliged to file

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 17 royalty returns. These are required to be filed every 3 months. The filing and corresponding payment must be done during the first 10 working days of the month following the period assessed. The La Colosa property is 100% owned and operated by the registrant, AngloGold Ashanti. The La Colosa project mining activities are suspended due to a force majeure condition, based on the lack of environmental permits to operate. As a result of force majeure, the concession contract obligations are suspended. Present, possible liabilities from previous exploration activities exist related to the rehabilitation of exploration drilling platforms. These liabilities are specified in previous temporary forest subtraction permits and amount to approximately $5.9m. The drill platforms must be "returned into the forest reserve" according to the terms established in the published temporary subtraction permits. 4 Accessibility, climate, local resources, infrastructure and physiography 4.1 Property description La Colosa is located on the eastern flank of the Central Cordillera of Colombia. Elevations range between 2,500 and 3,400m above sea level. Farming has occurred up to elevations of around 3,000m with high Andean Forest cover (including foggy forest) persisting to about 3,400m. Above that elevation the two canopy-forest gives way to scrubby bushes. Birds and small mammals are the most visible species. The Pan-American highway, the main route between the capital Bogota and the port of Buenaventura, passes close to the project. Access to the actual project site is only by 4x4 pickup trucks. Cajamarca, a small town of some 10,000 inhabitants and at 1,800m elevation, is situated some 8km SE of the project area. Average precipitation in the project area is around 1,400mm with rainy and dry seasons. Heavy rain results in a predictable swelling of the volume of creeks. The company has purchased farmland required for the project. Except for an island of land at the El Diamante sector, and sectors in the vicinity to the Pan-American highway, AngloGold Ashanti Colombia is the owner of the required land for project layouts. 5 History Greenfields exploration activities commenced in 2006 and led to an exploration drilling program starting in March 2007. The existence of a multi-million-ounce gold deposit was evident in mid-2007 and drilling activities were subsequently increased. The initial PFS studies commenced in September 2008. In 2014, the project team presented an update of the initial conceptual study and reframed the PFS study deliverables. A final update of the AngloGold Ashanti’s PFS standard was introduced in November 2020. The project as such delivered initial financial models, pit shell optimisations, metallurgical test work and plant designs, designs for waste rock and tailings storage and associated facilities, and designs for camps and roads. The project has undertaken both Mineral Resource and condemnation drilling between 2007 and 2017, all drilling was diamond drilling: • 2007 - 2008: 16,974m Mineral Resource drilling • 2010: 8,802m Mineral Resource drilling • 2011: 49,442m Mineral Resource drilling • 2012: 23,482m Mineral Resource drilling • 2013: 6,550m Mineral Resource drilling • 2014: 17,054m Mineral Resource drilling • 2015: 402m Mineral Resource drilling, 1,673m condemnation analyses on infrastructure geotechnical drill holes • 2016: 544m Mineral Resource drilling, 926m condemnation analyses on infrastructure geotechnical drill holes AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 18 • 2017: 516m Mineral Resource drilling, 23m condemnation analyses on infrastructure geotechnical drill holes The following geophysical studies have been carried out: • Between 2009 and 2011: IP Survey contracted with Arce Geofisicos (Peru), was carried out in three phases: o Phase 1 (November 2009 to February 2010), o Phase 2 (March to April 2010), and o Phase 3 (December 2010 to January 2011). • A total of 145,300 meters of 43 Induced Polarization profiles with constant-spacing measurements of 50m intervals along E-W lines, employing the Pole-Pole (2-Array) electrode configuration were completed. • In 2010: Ibague El Aceituno core shed: Electrical properties were obtained on drill core specimens using a SCIP™ (sample core IP tester) instrument. The half-core samples had been submerged in water for more than 1 month. • In 2011: A refraction seismic study survey was contracted with Arce Geofisicos (Peru) and covered the El Diamante plant site. • In 2012: The greenfields team carried out a regional helicopter magnetic and radiometric survey, which included the La Colosa project site. A total of 7,134 km of radiometric and gradient magnetic data were acquired during an eight-month period. • Between 2015 to 2016: Downhole geophysics was contracted with Weatherford. Well-logs (Caliper, Density, Full Wave Sonic, Acoustic Televiewer, and Gyro) were obtained from 25 infrastructure drill holes. Initial flowmeter studies were performed on four holes. • In 2016: Electrical tomography contracted with Hydroingenieria was carried out at Alto Girardot, a potential plant site, and at Site 11 for the potential conventional tailings deposition dam. Tomography data was collected on 6 lines totalling 4,600m. Drillhole and Geoprobe direct push information were included in the interpretation. La Colosa was discovered in 2006 as a -200 mesh stream sediment anomaly of 669 ppb Au in what is known now as the La Colosa creek. The follow-up surface rock chip sampling delineated an area of approximately 300m x 100m averaging approximately 1g/t. The initial Inferred Mineral Resource presented in the 2008 conceptual study was 13.1Moz gold at 1g/t at a gold price of $1000/oz and a cut-off grade at 0.3g/t. The follow-up PFS Mineral Resource drilling was mostly carried out between August 2010 and end-2014. Hydrogeological, mine geotechnical, and infra-structure geotechnical drilling were performed between 2014 and 2017. The Mineral Resource is delineated by an optimised shell at $1,400/ounce and is thus seen as economic, however, the project remains on care and maintenance. The 2017 Mineral Resource estimation was used for the 2020 and 2021 Mineral Resource statement and reports 28.3Moz @ 0.84g/t at a $1,400 gold price. The in situ Mineral Resource has not been restricted by the Los Nevados paramo boundary which was formally declared by the Environmental ministry (MADS) in December 2016. The reason being that the Los Nevados paramo boundary is being legally challenged by AngloGold Ashanti. In addition, the zonation mapping and the environmental management plan (involving local landowners such as AngloGold Ashanti) of the Los Nevados paramo have not proceeded. The Los Nevados paramo extends over four different local government departments and requires integrated work programs from the autonomous regional environmental authorities. The La Colosa project team completed a conceptual study and was ready to advance to PFS. The conceptual study scenario demonstrated that the La Colosa project has a positive business case and is technically viable. To this end, the project team needed to develop a PFS project charter. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 19 6 Geological setting, mineralisation and deposit 6.1 Geological setting The La Colosa project is located east of the continental divide on the crest of the Central Cordillera, in between the regional geological structures of the Romeral and Palestina fault systems. The greater than 1km wide Romeral fault system occurs some 6km to the west of La Colosa. The Palestina Fault system is located at the deposit. The Central Cordillera, a geomorphological feature, includes parts of the Romeral Melange, the Cajamarca Valdivia terrane, and the Ibague Block. The Cajamarca Valdivia terrane is made up of Paleozoic basement rocks and deformed and metamorphosed volcano-sedimentary sequences commonly described as Cajamarca Complex. These schists carry detrital zircons of Triassic-Jurassic age. The Jurassic Ibague batholith (Ibague block) intruded older basement rocks of the Central Cordillera. The autochthonous rocks of the Central Cordillera are separated from the accreted oceanic crust of the Western Cordillera by the Romeral fault system, a tectonic melange with different shear blocks bounded by anastomosing faults with predominantly reverse movement. The general trend of the fault zone is NNE-SSW, and the fault is many kilometres wide. The San Jeronimo fault is the easternmost expression of the Romeral Fault system and places the metamorphic rocks of the Cajamarca complex on top of the volcanic-dominant late Cretaceous Quebrada grande complex. The San Jeronimo fault is exposed near the western limit of EIG-163 concession, and the fault zone exposes a variety of lithologies including ultramafic and mafic rocks, basalts and pillow lavas, chert, black schist, greenschist, and quartz+mica+carbonate schist transitioning to impure marbles. The Palestina Fault, which is the most important structural geological component when considering that La Colosa is considered a split off from the San Jeronimo Fault. The Palestina Fault continues in a NE-SW oblique direction across the Central Cordillera. The mayor Andean orogeny occurred in the Miocene when the Panama block collided with the Western Cordillera. It gave rise to continued uplift of the Central Cordillera, which is still ongoing. This results in the formation of an intermediate to acidic arc which is related to the continued subduction of the Nasca plate to the west. The regional project area is characterised by multiple intrusive stocks of dioritic to tonalitic composition (Figure La Colosa regional geology and geochronological data). It is postulated that the magma chambers for these stocks must have had an extension of at least 10km to 15 km. The U-Pb zircon ages of the La Colosa porphyry stock range between 7.4Ma and 8.5Ma, indicating the emplacement of the early, intermineral and late porphyry phases occurred in a narrow time range of 1.1 million years. On a regional scale, geochronological age dates of the adjacent porphyry stocks to La Colosa show ages ranging from 6.3Ma to 8.4Ma indicating a well-defined cluster of magmatic activity. There are more porphyry prospects along the Palestina Fault with the porphyry occurrences becoming younger to the northeast. For La Colosa, there is additional Re-Os dating of molybdenite giving similar ages of around 8 million years, yet there are also younger ages of 6 and 3 million years. The recent volcanism (Northern Andean Volcanic Zone) includes the Nevado del Ruiz (also close to the Palestina Fault), the Nevado del Tolima, and the Cerro Machin volcano. The Cerro Machin volcano is located about 10 km east of La Colosa, from which historic eruptions account for the ash blanket in the project area. The project 1:5,000 geological map covers the actual La Colosa deposit site, the El Diamante infrastructure site, and the tangible infrastructure area 14C. The San Jeronimo fault, which is greater than 500m wide, and at the westernmost point of mapping, consists of lenses of various rock types surrounded by an anastomosing pattern of faults and shear zones. Most faults are orientated N-S with moderate to steep dips to the east. A porphyritic dyke, also crosscut by later faulting, has been emplaced into the San Jeronimo fault zone and indicates additional extensional deformation. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 20 The Cajamarca Complex forms a shallow synform against the San Jeronimo fault. The metamorphic sequence exposes meta-gabbros (amphibolites) at the base overlain by black-schist and greenschist. The metamorphic sequence in the overall deposit and infrastructure area is characterised by: • Major sets of foliation planes (s1, s2) strike N-S to NNE-SSW. Dips are variable and range from sub- vertical to moderately to the east (90-60 degrees); another set dips about 25 to the E and SE. West- dipping planes are less developed, indicating that the folds are slightly west vergent (inclined to the west). • Fold axes (b1) plunge 0-30 degrees to the N and S, varying from NNW-to-NNE and SSW-to-SSE. Fold axes (b2) are less common and plunge approximately 10 to 45 degrees to the E (NE-to-SE). • Ductile shear zones are orientated N-S to NE-SW, dipping to the E and SE; the dip varies from about 20 to 50 degrees and from 65 to 85 degrees. In particular, the shear zones with steeper dips often show an overprint of brittle deformation. At or close to ductile shear zones, quartz remobilisation, crenulation, and folding are more intense. • Brittle faults vary in orientation, however, NNW-to-NNE-trending faults with steep dips (70-85 degrees) to the W, NW, E, and SE predominate. La Colosa regional geology and geochronological data.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 21 La Colosa local geology La Colosa deposit geology AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 22 The stratigraphic sequence was affected first by regional metamorphism during the Andean Orogeny, and subsequently by thermal contact metamorphism caused by the emplacement of the La Colosa, the La Guala, Sincelejo, and El Diamante porphyry stocks. With the exception at La Colosa, only weak hydrothermal alteration events overprinted the original metamorphic textures and altered the schistose host rock (silicic, potassic, and sodic-calcic alteration). The La Colosa site contains an intrusive complex with two magmatic centres known as the La Colosa and San Antonio porphyry stocks, hosted by the schistose country rocks. The La Colosa porphyry Au mineralisation is hosted by a composite porphyry stock of dioritic to tonalitic composition. The porphyry complex consists of three intrusive stages: The early, the intermineral and the late-stage magmatic event. The complex is present over a map area of 3.5 km2. In comparison, the much smaller San Antonio porphyry stock 1.5km to the southeast of the La Colosa complex is made up mostly late intrusive rocks and hydrothermal breccias. The metamorphic sequence can be broadly separated into grey-black graphitic schists, pale quartz-mica schists. Greenschists occur mostly at the El Diamante infrastructure site. The sequence includes a regionally continuous stratigraphic layer of quartzitic meta-sediments. Hornfels after schist is well developed along the entire La Colosa ridge. The most obvious feature of ductile deformation, D1, in schistose rocks is a penetrative foliation, S1, which often lies parallel to the original sedimentary layering, S0. The S1 foliation generally strikes north-south (NW-SE to NE-SW) and dips steeply to the east and west. Fold axes B1 plunge sub-horizontally to moderately to the north and south. Brittle deformation (faulting, fracturing, and brecciation) is evident in the metamorphic basement rocks of the Cajamarca Complex as well as in the younger intrusive stocks of the La Colosa area. N striking faults follow pre-existing weaknesses at ductile shear zones and along fold structures. NE and NW striking brittle faults are the most prominent brittle structures. The main NE-to-NNE-striking structural feature is the regional Palestina fault system, characterised by a series of parallel faults, such as the La Ceja fault, the NE-00 boundary fault and the NE-3 fault. Based on drilling information, the prominent brittle faults consist of multiple fault and damage zones reaching a width of up to 250m. Kinematic indicators on the NE striking faults indicate predominantly left-lateral and minor right-lateral and normal movement. The N and NW striking structures, such as the La Colosa fault, are controlled and limited by the major NE striking strands of the Palestina fault system. Based on kinematic indicators, these faults display extensional and minor strike-slip characteristics. The geometry of the N and NW trending secondary faults and the NE trending branches of the Palestina fault system, together with the kinematic indicators describe a sigmoidal-shaped pull-apart structure. At the northern end of the La Colosa area a N-striking structural corridor consisting of a set of parallel N-striking, E- dipping extensional faults controls high-grade gold mineralisation. East-to-ESE and ENE-striking faults and fracture zones are common and represent late-stage extensional deformation, cross cutting the earlier structures. The multiphase diorite porphyry gold complex can be divided into three phases (early, intermineral and late) and is elliptical in shape with a known maximum north-south axis of at least 1,200m. The complex strikes N10W with a dip of 75 east-northeast, the contacts are mostly structurally bound. Intermineral and late dacitic dykes extend both north and south into the foliated schistose hornfels. The highest-grade gold mineralisation is closely associated with a suite of early porphyry intrusions/breccias with potassic and sodic-calcic alteration, high intensity of gold-sulphide veinlets and sulphur values generally exceeding 2.5%. La Colosa early porphyry The early intrusions occupy a surface area of approximately 0.35km2 and include five diorite porphyries (E0, E1, E2, EDM, and E3) and three early intrusion breccias (EBX1, EBX2, and EBXDM). Early porphyry E0 (La Colosa Petrography, Photo A), which is only observed in drill core, is characterised by 20 to 30 volume percentage of plagioclase and hornblende phenocrysts in a microcrystalline groundmass. Diorite porphyry E1 is equigranular (La Colosa Petrography, Photo B), fine-to-medium grained, with a crowded texture and 60 to 70 volume percentage of plagioclase, amphibole, and orthoclase phenocrysts. Diorite porphyry E2 is medium-to-coarse grained with 20 to 30 volume percentage of subhedral to euhedral plagioclase and hornblende phenocrysts (La Colosa Petrography, Photo C). The EDM porphyry is a fine-to-medium grained diorite with 40 to 50 volume percentage of subhedral to euhedral plagioclase, hornblende and biotite (La Colosa Petrography, Photo D). The last early diorite porphyry E3 is fine-grained, with 30-40 volume percentage of anhedral to subhedral plagioclase phenocrysts (La Colosa Petrography, Photo E). E3 intruded AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 23 the earlier porphyries and generated intrusion breccias, such as EBX1(with E1-dominant clasts; La Colosa Petrography, Photo F), EBX2 (with E2-dominant clasts; La Colosa Petrography, Photo G), and EBXDM (with EDM-dominant clasts) cemented by porphyritic E3 diorite (La Colosa Petrography, Photo H). La Colosa intermineral porphyry Two intermineral stocks have been distinguished, covering a total map area of over 0.33 km2 (Figure La Colosa deposit geology). The earlier of these, I1, is a porphyry with subhedral plagioclase and hornblende phenocrysts (La Colosa Petrography, Photo I). The later stock I2 is a diorite porphyry containing 20 to 30 volume percentage of subhedral to euhedral plagioclase and hornblende phenocrysts (La Colosa Petrography, Photo J). Crosscutting relationships show that I2 intruded I1 and thereby formed an intrusion breccia IBX (La Colosa Petrography, Photo K). The coarse-grained early-stage and intermineral diorite porphyry are difficult to distinguish due to their compositional and textural similarities. La Colosa late porphyry The late porphyry stage at La Colosa has a map extent of approximately 2.7km2 and is made up of diorite (LD), quartz diorite (QZD), and tonalite (TO) porphyries as well as intrusion breccias (QZDBX), some at the contact with country rock schist (DBX) (La Colosa deposit geology). The principal rock is the tonalite porphyry (TO) with phenocrysts of plagioclase, quartz, biotite, and hornblende (La Colosa Petrography, Photo l) in the northeast corner of the La Colosa Complex. The late porphyries are also exposed as NW striking, sub-vertical to E-dipping dykes cutting the early porphyry intrusions and schistose country rocks whole following existing faults, which formed with the pull-apart structure. San Antonio porphyry stock The San Antonio porphyry stock (La Colosa deposit geology) is divided into three late diorite porphyry intrusions (LS, IS1, and IS2). Their mutual contact and their contact with the schistose country rocks are marked by intrusion breccias (SABX) characterised by subangular clasts of wall rock, IS1, and IS2, cemented by late diorite porphyry LS (La Colosa Petrography, Photo S). The last phase of the San Antonio stock consists of hydrothermal breccias (HYBX) containing subrounded clasts of wall rock and minor diorite in a rock-flow matrix that is locally cemented by sulphides or carbonate minerals (La Colosa Petrography, Photo T). La Colosa rock petrography AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 24 6.2 Geological model and data density The exploration program was planned based on a typical porphyry gold model and with higher grade mineralisation being structurally controlled. The main components of the geological model are the 3D model of lithological units, the distribution of hydrothermal alteration and sulphides, and veinlet types and abundance. The 3D model of lithologic units shows the 3D configuration of all lithological units described above, based on surface mapping and core logging (La Colosa 3-dimensional (3D) lithology and WE section). A detailed metamorphic volcano-sedimentary stratigraphy can be obtained from hornfels applying classification schemes using high field strength elements, in particular V, Ti, Cr, Ni, Nb, P y Ce: • Horizon 1: Felsic, quartz-mica (biotite, muscovite) schists. • Horizon 2: Anomalous for Nb >30 ppm, P >2000 ppm, and Ce >100 ppm. • Horizon 3: Mafic, dark, actinolite-hornblende-chlorite-epidote schists (generally > pyrrhotite). • Horizon 4: Intermediate, quartz-biotite-actinolite (epidote) schists Distribution of hydrothermal alteration and sulphides: Eight, broad mineral assemblages of hydrothermal alteration are distinguished macroscopically: • Sodic-calcic (albite + actinolite + epidote); • Potassic (biotite + K-feldspar), which occurs mainly as vein infill and as pervasive replacement of wall- rock ferromagnesian minerals, such as biotite and hornblende; • Quartz-sericitic (white to gray mica + quartz), which results in partial destruction of the original rock textures; • Chloritic (chlorite alone), which is a late-stage alteration replacing biotite and amphibole; • Propylitic (chlorite + epidote + albite + calcite + actinolite); • Intermediate argillic (smectite + illite), which overprints the early alteration types, forming patches and selvages; • Silicification; • Supergene argillic Fe oxide alteration (kaolinite + iron oxides and oxyhydroxides), which is weathering showing box work textures after sulphide minerals. At the La Colosa intrusive centre, the predominant type of hydrothermal alteration in the early porphyry is moderately intense potassic alteration. This alteration type, with pyrite as the predominant sulphide, is locally restricted to the upper part of the early intrusive stage. The deepest members of the early stage, the EDM porphyry and EBXCM breccia are dominated by sodic-calcic alteration and magnetite veinlets with subordinate potassic alteration. The early-stage sulphides are pyrite, chalcopyrite, and molybdenite accompanied by hydrothermal magnetite. The intermineral stocks and breccias are dominantly altered by potassic (secondary biotite) and sodic-calcic assemblages with some cores of chloritic alteration. Pyrite is the most abundant sulphide, followed by pyrrhotite, which is commonly found close to the contacts with the country rocks. The dominant hydrothermal alteration types in the late porphyries are propylitic followed by intermediate argillic. Hydrothermal alteration of the San Antonio intrusions is predominantly propylitic with relics of secondary biotite and sodic-calcic alteration. In contrast, the hydrothermal breccias are dominated by intermediate argillic alteration. Disseminated sulphides include pyrrhotite and pyrite with the local occurrence of arsenopyrite, chalcopyrite, and molybdenite part of the hydrothermal breccia cement. Melnicovite (colloidal pyrite) has been identified as a breakdown mineral after pyrrhotite. The hornfelsed metamorphic rock sequence has been partially overprinted by hydrothermal potassic alteration that extends along foliation planes. Less abundant alteration types are several stages of silicification and localized sodic- calcic alteration. In contrast to the other alteration types, quartz-sericite alteration is clearly structurally controlled: it is spatially limited to an N-striking corridor of extensional faults, which crosscuts all lithologies. It overprints early potassic alteration in the early and intermineral porphyries and metamorphic rocks, as well as propylitic and intermediate argillic alteration in late porphyries. The N-striking faults are also locally affected by intermediate argillic alteration. Nine main veinlet types have been distinguished based on their mineralogy, textures and alteration halos. The most common veinlets are A and S types, and the earliest EB veinlets. A-type veinlets are characterised by vitreous or granular quartz along with sulphides, mainly pyrite, pyrrhotite, molybdenite and chalcopyrite, as well as magnetite. These veinlets (1mm to10mm wide) are also sinuous and occur with or without albite

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 25 + actinolite alteration halos. S-type veinlets (1mm to 5mm wide) are rectilinear, are dominated by pyrite and pyrrhotite, and occur in all rock types in the La Colosa stock and its country rocks. Early biotite (EB) veinlets consist of fine-grained biotite that fills sinuous veinlets (less than 2mm wide) associated with the early porphyries. No alteration halos occur along these veinlets. B-type veinlets (2 5mm wide) occur in rectilinear networks and are filled with granular quartz with central sutures lined with pyrite; they occur in potassic- altered early and intermineral porphyries, but they are scarce. M-type veinlets (1mm to 5mm wide) are sinuous or rectilinear and are irregularly filled with magnetite with or without actinolite, K-feldspar, pyrite, molybdenite, and chalcopyrite. Some of these veinlets are bordered by albite + actinolite halos. They occur in all rock types and are spatially associated with the E3 and EDM porphyries in the deep parts of the La Colosa porphyry stock. N-type veinlets (1mm to 5mm wide) are filled with albite accompanied by actinolite and sulphides. These rectilinear veinlets are bordered by sodic-calcic alteration, and they occur mainly in the early and intermineral porphyries. Actinolite (AC-type veinlets (less than 3mm wide) are rectilinear, contain traces of pyrite, pyrrhotite, molybdenite, and melnikovite, and are spatially associated with potassic and sodic-calcic alteration in the early and intermineral porphyries and in the schistose wall rocks. They present no alteration halos. D-type veinlets (1mm to 5mm wide) are rectilinear, filled by pyrite, and display fine-grained quartz-sericite halos. They are confined to the N-striking dilational corridor. Sheeted quartz veinlets (closely spaced and 1mm to 10mm wide) contain patches of albite + pyrite + visible gold + illite + carbonates, and locally they show crustiform-colloform banding and drusy quartz. The same mineral assemblage occurs as alteration halos around the veinlets. These rectilinear veinlets are confined to the N-striking dilational corridor. The project had difficulties obtaining permits for platforms in forested areas. Drilling in forested areas was restricted to forest openings along trails. Platforms at elevations greater than 3,200m were not accepted by the Environmental ministry in temporary forest extraction processes described under Environmental and Permitting). This affected Mineral Resource drilling intending to extend/better define high-grade mineralisation towards the north. The objective of Mineral Resource drilling during the PFS phase was to define the limits of the deposit, define higher grade areas for payback, and provide samples for metallurgical and geotechnical samples for test work. Initial greenfield's surface sampling (leading to the drilling proposal in 2006) depicted a coherent area of approximately 100m x 300m averaging around 1g/t. Drill holes COL001 and COL002 were placed in the centre of the greenfields anomaly and now form part of the high-grade mineralised envelope. Early environmental platform permitting allowed for a general 100m x100m drilling pattern. This was mostly confined to areas with grassy vegetation and defined an area between the La Colosa porphyry and the San Antonio porphyry. The limits of mineralisation were defined towards west, east, and south and partially at depth. Grade continuity for schistose rocks has been interpreted to be geologically controlled by foliation and contact zones to the diorite porphyry complex. Intercepts in schistose hornfels are more erratic and contain a higher amount of internal waste. Additionally, spotty high-grade also occurs outside the mineralised envelope. Oblique follow-up drilling was required in some areas, where initial grade continuity to form a mineralised envelope was missed or the geological interpretation appeared too optimistic. An internal metamorphic stratigraphy was modelled using ICP geochemistry data from the 100m x100m drilling. Presently, only three ore types are distinguished at the La Colosa porphyry deposit: Diorite, Schist, and Oxide with initial metallurgical test work existing for the San Antonio porphyry. Metallurgical questions arising from organic carbon (graphite content) could not be solved from the 100m x 100m drilling. The 1:1000 scale structural geological model was initially built from the 100m xm100m grid drilling and then refined with drill holes for Indicated Mineral Resource classification. The structural geological model was critical for the interpretation of the high-grade mineralised envelope and the interpretation of lithological contact between intermineral diorite and the schistose hornfels. Drill spacing of 75m is required for Indicated Mineral Resource classification assuming annual production of greater than 20mt per year. Many infill drill holes had to be drilled oblique or radially, as temporary forest clearing for platforms was confined to the approximate 100m x 100m grid. About 16 radial holes (Mineral Resource and geotechnical) were drilled from platform F5a, no additional Mineral Resource platforms could be permitted further north at the La Colosa ridge. The La Colosa deposit has not been drilled out to its limits, mineralisation is open to the north and at depth. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 26 The hydrogeological model, in general, ties in with the structural geological model, however, hydrogeological drilling is incomplete for the northern part of the deposit. The geological concepts of the La Colosa deposit have been published by the La Colosa geology study team led by the QP: Naranjo, A., Horner, J., Jahoda, R., Diamond, L.W., Castro, A., Uribe, A., Perez, C., Paz, H., Mejia, C., Weil, J. (2018): La Colosa Au Porphyry Deposit, Colombia: Mineralisation Styles, Structural Controls, and Age Constraints. Econ.Geol. 113, 3, 553-579. Selected conclusions from the publication are: • The regional NNE-striking Palestina fault system crosses the entire deposit area. The fault system is composed of several branches and intermediate splays. Like the adjacent Romeral fault system, the Palestina system reversed its slip direction from right lateral to left lateral due to a major change in the stress regime during the docking of the Panama-Choco block to South America in the mid- Miocene. The slip reversal reactivated older structures and formed new brittle, extensional structures. Within the regional Palestina fault system, the La Colosa deposit is located in a pull-apart structure that formed upon brittle reactivation of old ductile features and the generation of secondary Riedel shears. • This extensional zone favoured emplacement of the composite magmatic stock of the La Colosa intrusive complex. The complex is composed of three main intrusive pulses denoted early, intermineral, and late stage. The porphyry stock was emplaced during a short time span of approximately 1.1m.y. between approximately 8.5Ma and approximately 7.4Ma. • Potassic alteration is the dominant alteration type in the La Colosa porphyry stock. The schistose country rock also shows weak to strong potassic alteration accompanied by multiple stages of silicification. Pyrite is the dominant sulphide in the gold-rich (1.5-2.0g/t) zones of the deposit. Pyrrhotite forms an aureole in the country-rock. Close to the contact, pyrite and pyrrhotite coexist. Late-stage quartz-sericite alteration is associated with N-trending normal faults. • The early porphyry intrusions have the highest gold contents inside the 0.5g/t shell. This gold-rich zone lies above and overlaps a zone of magnetite, chalcopyrite, and molybdenite that is present in the deepest explored portions of the deposit. • Following emplacement of the stocks and the porphyry-type Au-Cu-Mo mineralisation, continued extensional deformation during uplift of the Central Cordillera formed a set of N-trending normal faults. These faults were sites for renewed hydrothermal fluid flow and deposition of the second, high-grade Au mineralisation within sheeted, drusy quartz veinlets bordered by sericite-albite halos. These veinlets crosscut the early, intermineral, and late porphyry intrusions as well as the country rocks in the northern area of the La Colosa deposit. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 27 La Colosa 3D lithology and WE section. 6.3 Mineralisation Mineralisation and metal zoning Metallurgical and geometallurgical studies on composited and selective samples indicate that gold occurs predominantly as native gold, as electrum, and in minor quantities as gold tellurides and gold-silver tellurides. Gold occurs as isolated grains and as inclusions or fracture fillings in pyrite, pyrrhotite, and silicate minerals such as feldspar and quartz. Three gold mineralisation events have been recognised at La Colosa: The first is associated with the various porphyry phases of the intrusive complex. The early porphyries and early intrusion breccias have gold ranging in grade from 0.75g/t to 1.0g/t. Veinlets of A and S type are abundant, and potassic alteration is common. The gold grade of the intermineral porphyries and breccias is lower, from 0.5g/t to 0.75g/t. In the late-stage porphyries, the gold grade drops to less than 0.3g/t. In this first event, gold was concentrated at the contact zone between the intrusions, and the intrusions and their country rocks, also in fracture zones that follow the NNW to N-trending D1 folds in the country rocks. These zones can be traced through the schistose country rocks for 1km north and south of the La Colosa porphyry stock. The second gold event is centred on the N- striking extensional faults that crosscut the early, intermineral, and late-stage porphyries as well as the schistose country rocks. Sheeted quartz veinlets are abundant, and they contain quartz and pyrite with centimetre-wide halos of albite + sericite + carbonate. In the late tonalite porphyry, this second event is represented by sparsely distributed pyrite + sericite veinlets (less than 3 per m in drill core) with elevated gold grades (less than1g/t). At La Colosa these high-grade sheeted quartz veinlets occur along the strike of faults for hundreds of meters into the northern part of the deposit, and their extent remains open for further exploration. The third gold event is a local product of supergene enrichment in the late tonalite and quartz diorite porphyries due to weathering that produced clays and iron oxides. In this, gold grade is enriched from 0.1g/t in unaltered rocks to greater than 0.5g/t in weathered rocks. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 28 Pyrite - pyrrhotite distribution Flotation concentrates have been prepared in metallurgical tests. Pyrite is the dominant sulphide followed by pyrrhotite. Small amounts of chalcopyrite are common. Marcasite, arsenopyrite, bornite, covellite, galena, and sphalerite occur as trace amounts. Pyrite - pyrrhotite overlap in their occurrence, however, proportions are generally inverse (See - Sulphide distribution based on detailed logging). The best mineralised sector (grade shell 1.5g/t and 2.0g/t) is characterised by greater than 5% pyrite in the core zone surrounded by an area of 3% to 5% pyrite. The wider area carries 1% to 3% pyrite. The mineralised saprolite east of the La Colosa creek overlays a narrow zone of greater than 5% pyrite. Pyrrhotite can crystallize in nature in the form of hexagonal and monoclinic crystals. At La Colosa there is both iron-rich, magnetic (hexagonal) pyrrhotite and iron-poor, non-magnetic (monoclinic) pyrrhotite. The early diorite in the centre of the porphyry complex contains for the most part less than 0.3% pyrrhotite. The intermineral diorite in proximity to the schistose country rock and the late porphyry sectors close to the schistose country rock accounts for the highest pyrrhotite contents (greater than 1.5%). The highest pyrrhotite contents have been recorded in the schistose hornfels between the La Colosa porphyry and San Antonio. There are sizeable areas with greater than 1.5% pyrrhotite. Porous, skeletal pyrite logged on drill core is low temperature, the colloidal form of pyrite, also known as melnikovite. Petrographic studies corroborate that melnikovite is a low-temperature alteration product of pyrrhotite. Melnicovite can later start re-crystallizing into marcasite. It appears that pyrrhotite and melnicovite contain less gold inclusions that pyrite. The gold grains observed in pyrrhotite-melnicovite recrystallization contacts remained stable. Skeletal pyrite (melnicovite) disintegrates in core boxes in less than two years into a salt-like material. The continuity of significant mineralised zones was interpreted using grade shells. The spatial occurrence of base metals in relation to gold mineralisation was interpreted from drilling, with additional fluid inclusion studies carried out on the subject. Gold-grade shells: The grade shells for 0.5, 1.5, and 2.0g/t illustrate the distribution of gold within the deposit. The 0.5g/t shell has been interpreted manually on sections and encompasses early and intermineral intrusions, and portions of the schistose country rock. Three separate zones can be recognised within this shell: • Zone 1 shows grades between 0.5g/t and 0.7g/t and is associated with porphyries E3 and EDM and the breccia EBXDM. All of these rocks exhibit potassic alteration and contain A- and M-type veinlets with pyrite and trace chalcopyrite and molybdenite. Gold zone 1 partially overlaps with a grade shell that contains the highest values of copper (greater than 0.1%), molybdenum (greater than 0.01%), and silver (greater than 1g/t) at La Colosa. • Zone 2 within the 0.5g/t shell is hosted by early porphyries (E0, E1, E2, and E3) and breccias (EBX1 and EBX2) and a portion of intermineral porphyry. Disseminated pyrite is the dominant sulphide. A, S, and D-type veinlets occur within the zone and uncommonly contain magnetite, chalcopyrite, molybdenite, and K-feldspar. The gold ranges in grade from 0.75g/t to 1g/t. • Zone 3 is marked where the abundance of pyrrhotite exceeds that of pyrite. In this zone intermineral porphyries and intermineral intrusion breccias predominate, showing potassic and subordinate chloritic alteration. Potassic and subordinate chloritic alteration is dominant in the central parts of the intermineral porphyry where the gold grade drops from 0.7g/t to 0.5g/t. A portion of the gold mineralisation in the 0.5g/t shell occurs in the hornfelse schistose country rocks. The mineralised hornfelsed schists display silicification, potassic, and sodic-calcic alteration, contain A and S- type veinlets with sulphides being pyrrhotite, pyrite and melnikovite. The 0.5g/t shell contains 1,107Mt of ore with 684t (22Moz) of gold averaging 0.62g/t. The grade shells for 1.5g/t and 2.0g/t are implicit grade shells constructed in Leapfrog™ software. These grade shells highlight the sectors above normal porphyry grade mineralisation. Both early, intermineral and late porphyries and intrusion breccias, as well as schistose country rocks, are the host lithologies for high- grade mineralisation. The 1.5g/t grade shell extends over 1.2km of surface length and contains 196Mt of ore with 271t Au (8.7Moz) at 1.38g/t. The 2.0g/t grade shell extends over 0.75km and contains 40.7Mt of ore with 84t Au (2.7 Moz) at 2.04g/t.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 29 Deep copper, molybdenum, and silver zone: Copper, molybdenum, and silver are closely associated in space, and their highest grades are found in the eastern part of the deposit between 2,400 and 2,600m elevation. The deep copper, molybdenum, and silver zone is hosted by early diorite (E3, EDM, EBXDM), with average grades being 0.11wt % Cu, 0.017wt % Mo, and 2.3g/t Ag. La Colosa grade shells. Sulphide distribution based on mapping and detailed logging. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 30 7 Exploration 7.1 Nature and extent of relevant exploration work The La Colosa Exploration history is outlined below: 2004: • AngloGold Ashanti Colombia commences the first stream sediment program in the middle Cauca region. 2005: • Five porphyry Au-Cu projects were identified (La Mina, Quebradona, Yarumalito, Orofino, Quinchia). 2006: • AngloGold Ashanti Colombia launches exploration in the Central Cordillera of Colombia between the towns of Salento and Cajamarca (part of the Ibague-Mariquita Project Region). • Dec. 2006: Drill proposal at La Colosa 2007 – 2008: • March 2007 - February 2008: First diamond drilling campaign (COL001-COL059) • September 2008: Start-up of La Colosa PFS project 2010 – 2014: • IP and seismic ground geophysics over mineralised zones and infrastructure sites Second Mineral Resource drilling campaign (COL060 COL397), Mineral Resource drilling officially ends. Geotechnical drilling, hydrogeological drilling, and installation of the groundwater monitoring system at the pit site. • 2014: Conceptual Study update 2015 – 2017: • Infrastructure geotechnical, hydrogeological, and borrow material drilling. Installation of vibrating wire and open pipe piezometers for Site 11 infrastructure designs. • Downhole geophysics. 2016: Geometallurgical block model for comminution included in the Mineral Resource block model 2016: Last update of the structural geological model 2016: First LUC block model 2016: Geophysical program: electrical tomography for infrastructure sites 2017: Golder Associates: Conceptual hydrogeological model for Site 11, Factual Geotechnical and Hydrogeological Report for 2015 - 2016. 2017: • Project enters care and maintenance 2021: • Inhouse desktop studies • LUC block model for 25m x 25m x10m, 12.5m x 12.5m x 10m, 10m x 10m x 10m SMU. • Update of geotechnical domain model. • Colosa Project 2.0 Alternatives Assessment Desktop Review. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 31 • Integrated Colosa block model for 10m x 10m x10m SMU, Mineral Resource classification for 15Mt annual throughput, sulphur ARD, and comminution. A total of 148,062m has been drilled to date. Three compliance drill holes (800m) and one geotechnical- hydrogeology drill hole were completed in early 2017 before activities were suspended. Studies on the deposit include: • R.H. Sillitoe reviewed early exploration data and his view accounts for the initial mineralisation model. • Master theses on Colosa were prepared at Rhodes University (SA), University of Arizona (USA), CODES/the University of Tasmania (AU), and University of Barcelona (E). • The QP and his geology team published the available chronological data of the project. • Orefind (Stuart Halley) prepared a report on La Colosa Geochemistry. The lithological wireframe model is based on the integration of outcrop mapping and diamond drill data. Outcrop mapping in the pit area is 1:1000, and for the surrounding infrastructure area 1:5000. Drill holes/sections are 100m for the pit area, and several 100m for the infrastructure area. Similarly, the structural geological model is 1:1000 scale for the pit area and 1:5000 scale for the surrounding infrastructure area. The structural geological model was subcontracted to Johannes Horner (iC Consulenten). The project database includes XRF and XRD analysis. Geometallurgical studies related to comminution modelling focused on obtaining hardness parameters have been undertaken while additional metallurgical comminution tests have been carried out for poorly represented areas. This metallurgical data has been correlated with multi-element assay and spectral mineralogical data to obtain proxies for metallurgical parameters. Some 43,529m (153 drill holes) have been spectrally scanned using a sisuMobi™ system equipped with a red-green-blue (RGB) camera and a shortwave infrared camera. The sample dataset includes the required metadata. An area of approximately 9km x 6km was covered by exploration and comprises the mineralised porphyry centres, the infrastructure area, and possible sites for construction material. Three 1:5000 geological map sheets are available for the Colosa study (La Colosa local geology). The sheets are named Area 14C, El Diamante, and La Colosa: • Geo_A0_SK_01: Area 14C • Geo_A0_SK_02: Diamante • Geo_A0_Sk_03: La Colosa Thematic maps (Au, Ag, As, Cu, Mo) have been prepared for the La Colosa resource sector. The area south of the Bermellon river has only been covered by regional 1:10,000 mapping (Naranjo, 2015). With exploration potential seen as low, it has been considered for infrastructure location. The Colosa database was created in 2006, initially for geological logging and assay, and then expanded (see database discussion). Sample preparation and analysis are conducted according to standard industry procedures. For sampling diamond drill core is cut in half, crushed, split, and pulverised prior to analysis (see Sample preparation). Gold is determined by fire assay and multi-elements by CCP-AES and ICP-MS after four acid digestion, both of which are total methods (see assay discussion). Drill samples are analysed as follows: Gold: Fire assay fusion with atomic absorption spectroscopy on 50g nominal sample weight (ALS code AU- AA24), using a gravimetric finish (ALS code AU-GRA22) for samples over 10ppm. Multi-element: analysis is conducted according to ALS method ME-MS61, which comprises a four-acid digestion (F-HNO3-HClO4 digestion, HCL leach), followed by elemental determinations by inductively coupled plasma atomic emission spectroscopy (ICP-AES) and inductively coupled plasma mass AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 32 spectrometry (ICP-MS). Over limit samples use atomic absorption method on individual elements (OG62). High S concentrations (over 10%) are determined by LECO. Mercury: For most samples, Hg was determined by cold vapor AAS after aqua regia digestion (ALS method Hg-CV41). However, from borehole COL305, Hg is determined by ICP-MS, ALS method Hg-MS42. The database backups (when the project was in operation) were performed using remote fusion with the SQL server located at the El Aceituno core shed. The following backups were made: • A differential backup, (only modified elements) was done every hour. • A full daily backup was kept throughout the current month. • At the end of the month, only a full backup of the last day was kept into the next month. The database was moved to the AngloGold Ashanti Colombia offices in Bogota when the El Aceituno core shed was put on care and maintenance. A check-in of all the La Colosa information to the central Bogota database was performed. Fusion™ remote is 100% compatible with the central fusion data structure. The data is stored in the SQL Server there and backups continue under the same protocol (differential backups and daily full backups). A copy of the latest version of the La Colosa 2017 Century Systems™ database exists at the AngloGold Ashanti Medellin office, the AngloGold Ashanti head office in Johannesburg, and in the data room organised by the QP. La Colosa is an AngloGold Ashanti greenfields discovery, and no outside data is used. 7.2 Drilling techniques and spacing All drilling is diamond drilling. Drilling was carried out by man-portable drill rigs in PQ-HTW, HQ-NTW, and NQ diameter using a standard inner tube for drill core recovery. Geotechnical drilling, structural geological logging, and the study of veinlet orientations for Mineral Resource drilling include orientated drill core. Orientated core drilling was obtained using a Reflex ACT II RD tool. Reflex provided training for both the drillers and the company platform technicians. For that purpose, the oriented core was pumped out of the standard inner tube. Kenometers were used for logging orientated core. All drill core has been logged and studied in detail. Logging manuals were created for Mineral Resource and geotechnical logging, and these include: • A manual for geological logging • An AngloGold Ashanti core logging geotechnical procedure • A Dempers core logging geotechnical manual • A protocol for logging soils Drill core has been measured according to the following protocols: • Point Load Testing (PLT) protocol • Density: using the Archimedes principle of the mass of a core sample in water and in air • Density paraffin method (weighing the original and the wax-covered sample in both air and water • Equotip test protocol to measure rock hardness • The sonic logger test protocol to measure drill core competency • The magnetic susceptibility test protocol For geological logging, drill core was logged visually for lithology, hydrothermal alteration, and porphyry veinlets. The logging of hydrothermal sulphides and oxides is quantitative, while the logging of indicative hydrothermal silicate minerals is semi-quantitative. The critical drill core has been scanned at the El Aceituno core shed. The hyperspectral short-wave infrared (SWIR) core imaging provides information on the mineralogy and the context in which that mineralogy occurs. Certain minerals have characteristic absorption features in the specified spectral range. The feature extraction can be used in quantifying mineralogy, displaying compositional trends of specified minerals

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 33 across the deposit, and facilitates geometallurgical correlations. The hyperspectral data was acquired using a sisuMobi system. The contractor was Geospectral Imaging (GSI), now Terracore™. Photos were taken of both wet and dry core. Some 148,062m of drill core has been logged for geology, hydrothermal alteration, and core recovery. Initially, a Pajari™ equipment and later a Reflex™ tool has been used for downhole survey with that corrected for magnetic deviation. The results of all downhole borehole surveys are recorded and then saved in the drill hole database. Category Spacing m (-x-) Type of drilling Comments Diamond RC Blasthole Channel Other Measured - - - - - - - Indicated 75x75 Yes - - - - Hornfels in ENV05 straddles at the limit between Indicated -Inferred Inferred 100x100 Yes - - - - Deepest drill hole: 1209.06m, multi-direction diamond drilling Grade/ore control - - - - - - - 7.3 Results The material results of the Mineral Resource drilling and their interpretation are explained using two examples filtered from the hole file created in Datamine™ software. Drillhole COL133 represents a porphyry drillhole, drillhole COL263 represents a schistose hornfels drillhole. The database query is shown on 15 columns: BHID (drill hole identity), FROM (sampled from), TO (sampled to), LENGTH (sample length or interval), SAMPLE_N (database sample number), Au_ppm (fire-assay results), AG_ppm, AS_ppm, CU_ppm, MO_ppm, S_ppm (ICP and Leko analysis for silver, arsenic, copper, molybdenum, and sulphur, only a short list of selected elements is shown), DENSITY (air-water and paraffin density estimates), RECOV_m, RECOV% (drill core recovery in meter and in percent), ROCK (logged rock type). The additional two columns Lith_wf and ENV were added for the purpose to demonstrate their interpretation. The Lith_wf column refers to the indicative 3D lithological wireframe and the ENV column refers to the indicative gold grade shell the sample interval relates to. Specifically, for drill hole COL133: The first drill run from 0m to 1.8m is characterised by significant core loss (SGNCRLSS) typical of built-up material (soil and volcanic ash) on the platform. The interval was assigned to the ash wireframe and was not sampled. The drill interval from 1.8m to 9.48m was logged as intermineral diorite breccia and from there onwards till 129.4m as Early Diorite 1. The lithological contact between the intermineral diorite breccia and early diorite occurs in partly weathered and, consequently, locates in the oxide-transition wireframe. The contact between oxidized rock and all-sulphide rock occurs at 28m according to 3D wireframe modelling. Drill core sampling has been as systematic as possible and the lithological contact at 9.48m was not taken into account. Sample 10077213 runs from 8m to 10m depth. DatamineTM software, nevertheless, will split the sample interval at lithological boundaries marked in the lithology input file. Datamine software will assign to both sub-intervals the same sample number and the same analytical result. A composite table for assay results was created for each drill hole. For the specific example of COL133, the entire drill core forms one composite: The intercept from 1.8m to 598m averages 1.48g/t. COL133 intercepts the deeper copper and molybdenum anomaly at around 480m. Copper values are generally above 1000ppm Cu, molybdenum analyses above 100ppm Mo. The copper-molybdenum anomaly partially overlaps with the ENV15 grade shell (displaced to the east at depth). Specifically, for drill hole COL263: The initial two drill runs recovered strongly weathered material logged as Saprolite (SAPR), the interval is assigned to the oxide wireframe. The geological logging manual does not distinguish between different saprolite rocks. The detailed logging of weathered material was carried out under geotechnical (Dempers) logging. The contact between oxide-transition and all-sulphide material occurs at 60m. Drillhole COL263 intercepts schistose hornfels (H), a narrow late dacite-tonalite dyke was logged between 280.52m and 288.4m. The real thickness of the dyke was obtained during wireframe modelling, which for this project was carried out in Leapfrog software. Similarly, as above, systematic drill core sampling occurred across the hornfels dyke contact. Compositing for gold to construct the explicit gold grade shell ENV05 indicates four composites above 0.5g/t: 0m to 136m @ 1.14g/t, 138m to 250m @ 1.36g/t, 300m to 316m @ 0.60g/t, and 320m to 342m @ 0.54g/t. COL263 represents one of the longer intercepts in schistose hornfels. Gold grades in schistose hornfels are more erratic than in the porphyry. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 34 The La Colosa drill hole database to March 2017 contains 456 drill holes totalling 148,062m of diamond drilling. Drill platforms are shown on the plan view with drill locations. There are 62,107 samples with gold assay and ICP-MS analyses. The difference between samples and (logging) records refers to geotechnical, hydrogeological, and specified metallurgical drilling. Drill holes shown in red have both Au-ICP analyses and geometallurgical proxy results. The La Colosa drill hole database includes 1,525 samples taken for condemnation purposes from 43 infrastructure drill holes. Mineral Resource drilling is centred on the La Colosa ridge. The ridge strikes NE-SW at its highest elevations, turns N-S and then NW-SE. The southernmost portion named San Antonio was the site of historic formal mining in the 1980´s. Drill holes COL133 and COL263 are marked in stronger black. The figure also shows a table for exploratory data analysis of Au, Ag, Cu, Mo, As, S and Density. Depth of drilling is depicted on a NNW-SSE section. The porphyry intercepts occur mostly over a vertical interval between 2400m - 3200m elevation, and the hornfels intercepts between 2,250m – 2,900m elevation. 7.4 Locations of drill holes and other samples Plan view with drill hole locations. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 35 Drill hole and exploratory data analysis. 7.5 Hydrogeology In the hydrometry database, there is information from 13 pressure transducers that measure water levels in the creeks to calculate water flow, information from 14 rain gauges, and three weather stations. Average rainfall for the La Colosa, the La Arenosa, and the La Guala basins are in the order of 1,500mm/year. The stations register data according to the methodology of IDEAM (Institute of Hydrology, Meteorology and Environmental Studies of Colombia). Water flow measures for creeks were carried out between 2012 and 2016, rain gauges were active between 2008 and 2016. The pressure transducers and rain gauges were stored at the camp when the project was stopped, the weather stations at the project site were abandoned. Groundwater levels and hydraulic heads were compiled from 51 piezometers. Thirty two are open hole piezometers and 19 vibrating wire piezometers in fully grouted boreholes. Similarly, that monitoring came to a halt when the project was stopped, and the care and maintenance phase started. The condition of the piezometers is not known. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 36 Location of hydraulic tests, piezometers, flow stations, rain gauges, and weather stations. Precipitation at La Colosa To characterise the hydraulic conductivity of geological units 448 hydraulic field tests were carried out. Some 191 tests were conducted in the Mineral Resource area and 257 in designated infrastructure locations, of these 384 are Lugeon and 64 Lefranc tests. Schlumberger Water Services Peru (SWS) was contracted in 2011 and 2012 to do hydraulic field tests in the mineral deposit area. INGETEC, a Colombian engineering consulting firm, carried out the hydraulic field tests for infrastructure locations during 2015 and 2016. The methodology for calculating the hydraulic conductivity follows the flow equation published by the US Department of the Interior Bureau of Reclamation Earth Manual, Part 2, p.1255 (third edition1990). The tests were audited and verified by AngloGold Ashanti’s technicians. Based on the general characteristics of the main lithological units and the hydraulic conductivities obtained, four hydrostratigraphic units were defined in the conceptual hydrogeological characterisation report (Golder, 2017).

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 37 The hydraulic conductivity values for the hydro stratigraphic units are: Hydro stratigraphic Unit Hydraulic Conductivity (m/s) Minimum Maximum Quaternary Deposit 4.2E-08 6.3E-05 Weathered Rock 6.4E-09 2.0E-04 Metamorphic Unit 1.0E-11 1.7E-05 Igneous Unit 1.0E-11 1.4E-05 Average annual recharge values were obtained using a selected tank model. The tank model is widely used to characterise the water balance in different regions and climates. The model considers the hydrological processes of precipitation, surface runoff, infiltration, and evaporation. The selected tanks represent the storage of water in capillary retention, direct runoff, subsurface, and base flows. On average, the recharge at the basin scale varies between 0.001 to 0.021 m3/s (Golder, 2017). 7.6 Geotechnical testing and analysis The geotechnical understanding of the La Colosa deposit is supported by detailed geotechnical studies conducted on diamond drill core data. Detailed geotechnical logging information was used in the definition of geotechnical design sectors, as well as the characterisation and classification of the rock mass at the La Colosa ridge. The acquisition of geotechnical information at the La Colosa ridge began in 2011. The first drill hole with geotechnical logging was COL142. Some 29 orientated drill holes were specifically planned for geotechnical reasons. The information was later complemented with geotechnical logging and sampling of Mineral Resource drill holes, hydrogeology drill holes, and specified drill holes for metallurgical studies. Thus, there are 89 drillholes (30,189.12m) with geotechnical information for the pit area in the present geotechnical database. All geotechnical logging for the La Colosa deposit followed the Laubscher and the Dempers guidelines, which complies with the ISRM methods to characterise the rock mass. A Kenometer device was used to measure dip and azimuth of fractures and other structural geological/mineralisation characteristics on orientated drill core. That information in combination with identified detailed fault zone logging was used in the generation of the structural geological model and, subsequently, in the construction of the geotechnical 3D domain model. A quality control step was performed on the geotechnical logging information before entering the data into the project’s database. The database runs a script and/or configures the variables to perform rock mass evaluations by different methodologies such as RQD-RMR Laubscher, RMR Bieniawski, and Barton’s Q- slope method. The information was complemented with results obtained from geomechanical tests performed on selected drill core samples. Those geomechanical tests were carried out to gather information on the resistance of different rock material and to study their behaviour according to modelled geotechnical domains. The geomechanical tests were performed by different laboratories located in Bogota, Medellin, and Tucson (Call and Nicholas, Arizona, USA). The table summarizes the type of tests, the number of samples tested, the number of samples available for the PFS , and the laboratories contracted. Test Type Quantity PFS TOTAL Laboratory UCS 298 298 University of Los Andes, Geomecánica Integral Laboratory, Medellin, Colombia Direct Shear 178 178 University of Los Andes, Call & Nicholas AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 38 Triaxial Compression* 110 110 University of Los Andes, Geomecánica Integral Laboratory, Medellin, Colombia Indirect Tensile 219 219 University of Los Andes, Geomecánica Integral Laboratory, Medellin, Colombia TOTAL 805 805 * Considers three triaxial tests with different confinement pressures The earlier PFS geotechnical work domained the hard rock units into Intrusive (generally closer to surface) and Metamorphic (generally lower down) domains with sub domains selected based on Rock Mass Rating. The data indicates that the intrusive rock is consistently hard with the metamorphic domains varying from very soft to medium hard. A number of tests have been carried out on discontinuities however these as well as the additional intact rock strength tests following the earlier geotechnical PFS study have not yet been assimilated into design parameters for future use. The design work to date was carried out to PFS level however due to pit boundaries changing and the additional data collected, the earlier PFS is of limited value. Consequently, further design analysis will be needed. It is expected that future geotechnical design work will focus on a re-interpretation of rock mass and discontinuity shear strengths, and analysis will focus on complex structure driven failure mechanisms based on the large number of known faults as well as rock fall which should include practical batter berm configurations. This latter part is particularly important as steeper hard rock inter ramp angles are very sensitive to batter scale structural failure mechanisms and blasting interaction. In summary, the rock mass is very competent, however the larger structures and their exact location with respect to pit boundaries will determine the failure mechanisms that drive slope configurations. Geotechnical drill holes across La Colosa AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 39 8 Sample preparation, analysis and security 8.1 Sample preparation Pre-project greenfields sampling during 2006 consisted of -200# stream sediment samples, soil samples and rock chip samples. The sampling followed the AngloGold Ashanti greenfields sampling protocol. Sampling from 2007 onwards consisted mainly of diamond drill core and minor Shelby sampling for volcanic ash and oxide. Surface rock chip samples included in the Mineral Resource model are rock chip and channel samples from an old, small-scale mining tunnel and a silicified rock exposure in the San Antonio sector. The Mineral Resource model does not include greenfield's surface and soil samples because of the difficulty to correctly place the samples in the oxide, transition, soft sulphide wireframes. All samples were prepared and analysed at commercial laboratories. Greenfields diamond drilling during year 2007 and 2008 was HQ and NQ core diameter using two drilling contractors. PFS study Mineral Resource diamond drilling commenced in August 2010 and was carried out by a single contractor (Kluane Colombia) using man-portable drill rigs. Drill core was obtained in PQ, HTW, NTW diameter. The depth capacity of drill rigs was between 200m and 1200m for NTW (NQ thin wall) diameter. A technician contracted by AngloGold Ashanti Colombia was responsible for receiving the drill core at the platform and the subsequent core box handling. The QP was responsible for establishing the site-specific logging and sampling protocols. Core recovery and RQD were measured at the drilling platform. During pre- logging at the platform or at the La Colosa camp rock type, veining, alteration, and mineralisation were identified and reported daily for guidance. The detailed Mineral Resource core logging was carried out at the El Aceituno core shed, and the core has generally been sampled at 2m lengths. Photos were taken before sampling and cutting the core. The soil horizon and the post-mineral ash layer were not sampled. Each half- core sample was packed in double plastic bags and eight-digit sample code tags were inserted. The samples were sent to the selected ALS Chemex sample preparation laboratory under the custody of the security department. The remaining cut diamond core has been stored at the El Aceituno core shed together with the returned ALS Chemex coarse rejects and pulps. The La Colosa database was created in 2007 to store data from the first drilling campaign (59 drill hole data). During the initial period, only geological logging and assays data were recorded. With the start of the PFS in 2009, the database has been growing to accommodate the requirements of Mineral Resource drilling and logging, geotechnical, geometallurgical, metallurgical and hydrogeological studies. The amount of information stored has substantially increased. QAQC procedures have been developed to ensure data quality. More complex and new QAQC methodologies will be needed to control future data entries once the project restarts in the PFS and FS stages. The La Colosa database has been developed using Century Systems software (later to become Fusion Database which is part of the Datamine Software suite). The database has been structured by subjects: SUBJECT SUB SUBJECT ITEM COUNT (meters) VALID/NO VALID DATA Geology Core Logging 148062.04 Assays Data 62107 XRF Data 132 Density Data 70800 Petrography 112 Infrastructure Core Logging 614.4 Weatherford Geotech Core Logging 38365.12 Geophysic Structures (WTF) 5163 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 40 Triaxial Test VALID / NO VALID 332 / 325 Direct Shear VALID / NO VALID 178 / 314 TI VALID / NO VALID 219 / 0 UCS VALID / NO VALID 299 / 172 Full Wave Sonic 34845 Geometallurgy Comminution Equotip Data 62043 Sonic Logger Data 43311 Magnetic Suceptibility Data 59169 BMWi Modif 457 A*B RBTLite 269 Proxy 68157 Spectral 200225 Recovery LeachWell 1022 Organic Carbon 5016 Mineralogical Data QXRD 1093 MLA 17 Metallurgy Comminution BMWi Full 42 A*B DWT 2 A*B SMC 8 SPI 273 CI CEET 237 CI CWI 8 HPI 32 TA DWT 2 TA SMC 7 Recovery Bottle Roll 96 / 224 NO VALID / VALID LeachWell 36 / 131 NO VALID / VALID BLG 643 36 / 131 Hydrogeology Piezometer Data 425891 Water Point Data 4878 Artesian Wells Data 1498 Lefranc 56 Packer 394 ARD Water 20 The hydrology database was managed in the past by the Environmental department. For quality assurance, the hydrology database requires software updates, internal reviews, and an external audit. The ARD data was neither compiled in a final report nor is it held in a systematic database. In 2021, the QP compiled most of the existing data into the company Sharepoint system. The database appears to be complete except for certain analytical sheets. The ARD data consists of static and kinetic studies on samples considered waste. The samples were mostly coarse rejects or broken drill core, minor surface samples, and only one set of samples from initial metallurgical test work.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 41 Gold mineralisation in the porphyry phases is considered disseminated and controlled by veinlet type, veinlet density, and sulphide content. Higher gold grades occur with increasing pyrite content and increasing porphyry-style (A, S veinlets). The geometry of gold mineralisation in the schistose hornfels has been interpreted to follow foliation. The late higher-grade structural controlled mineralisation (cross-cutting all lithologies) generally strikes NS-N10W/SSE steeply dipping to the east. The porphyry intrusions and contacts to schists generally dip 75 to the east. Foliation in schistose rocks follows that trend. Drilling, in general, was carried out on W-E sections, however, there is also drilling oblique to the sections or radial drilling. Drill holes for the W-E sections have azimuths of both 90° and 270°, the La Colosa ridge was drilled from both sides intercepting lithological contacts at variable angles. Certain drill intercepts for the higher-grade late mineralisation were intercepted at shallow angles because of the limitation on establishing drill platforms. Some of the post-mineral faulting with gouge sub-parallel to mineralisation caused severe drilling problems. This was taken into account in the geological model and top cuts were applied to high assay results. AngloGold Ashanti Colombia purchased the El Aceituno property in Ibague for drill core handling, logging, and sampling, core scanning, geometallurgical tests, and drill core storage. A soil testing lab for infrastructure studies was set up. The geology team operated the core shed until the project went into care and maintenance in April 2017. Site security maintained their administrative office at the core shed. Core boxes with cut drill core, non-cut orientated core, and geotechnical core remained at the core shed (unless core material was used for metallurgical test work). Sample rejects have been stored in containers at the core shed. Pulps have been stored on shelves at the core shed. The legal department operated the project after the project shut down due to care and maintenance. The El Aceituno core shed was then also used as a storage facility for office equipment, environmental instrumentation, and others. No maintenance of the core boxes has been carried out since April 2017; the core boxes are affected by wood moths and are deteriorating. A certain amount of oxidation affected the drill core, especially the drill core containing pyrrhotite. Reject samples were stored in a container and pulps on the shelves at the core shed. Geotechnical samples when returned from contractors were returned to the core shed and kept in separate core boxes. The material tested for construction purposes was kept in bags. A set of metallurgical samples for flotation fine-grained variability test work were kept in drums. On the drilling platform, core recoveries were obtained from the difference between drill run and core recovered in the inner tube. Volcanic ash and interlayered soil were mostly recovered dry by pushing and slowly rotating the drill rods. Colluvial blocks within the volcanic ash had to be drilled. Both the drilling company and the platform technicians were trained in mud programs to maximise core recovery under difficult drilling conditions. Specific muds at correct pH were prepared to recover ‘Weathering 1’ which is mostly crumbly material. Similarly, specific mud mixes were prepared for swelling clays in fault gouge intercepts. Nevertheless, a few drill holes were abandoned with rod strings left behind by the drilling company. Lubricants and specific additives were added to keep friction low. Unfractured porphyry and unfractured hornfels could mostly be drilled with water only and additives were used to lift detritus. There is no correlation between grade and recovery. Core recovery for sulphide material is generally greater than 90%. The fractured drill core was wrapped with masking tape after detailed logging which helped prevent the core from disintegrating during cutting. Care was taken that both the coarser and finer fraction was sampled in faulted drill core. Diamond drill core was generally sampled at 2m intervals without changing the 2m sampling interval at lithological contacts (different diorite pulses or at dike contacts in the schistose hornfels). The core has been cut with a diamond saw, only half core was sampled for assay. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 42 The sample preparation and analysis protocols were created by the AngloGold Ashanti Colombia greenfields team, audited by AngloGold Ashanti Head Office, and the laboratories tested by annual Round Robin tests. All required sample preparation and analysis steps were in place from the onset of the project. 8.2 Assay method and laboratory Sample preparation and assay were contracted to ALS Chemex. The sample preparation was carried out in their Medellin and Bogota laboratories, the assay and ICP analyses were done at their Lima laboratories. ALS Peru S.A. maintains a certificate of accreditation issued by the Standards Council of Canada for ISO/IEC 17025:2005. The initial registration date was 2010-03-01. SGS facilities were used to assay control samples. Sample preparation and analysis were conducted according to standard industry procedures and did not change over the project. The laboratory and QAQC reports were updated on an annual basis which is documented in the report Analysis La Colosa_Conceptual Report_Junio 2017 by Monica Uribe (2017): Routine sampling is based on 2m half drill core. Sample preparation following sample submission and reception at the commercial ALS laboratory consists of: • Drying at 110°C using stainless steel drying pans • Crushing to more than 70% less than 2mm using a terminator crusher • Splitting to 1 kg using a riffle splitter • Pulverizing to more than 85% passing 75 microns using an LM2 mill. • Splitting 250 grams to be sent to the analytical facility Analyses • Gold: Fire Assay and Atomic Absorption Spectroscopy on 50g nominal sample weight, with gravimetric finish for samples over 10g/t. Assay results were validated against detailed logging. Only very minor instances of sample handling errors (sample mix up, samples dropped or replaced) were detected, and then any suspicious intercepts or samples were re-submitted for analysis. • Multi-element analyses based on four acid digestion and ICP analysis • Leko analysis for elevated sulphur and sulphur species. • Leko analysis for total carbon and carbon species. Quality control procedure • Coarse blanks (two inserted at the beginning of the hole, and then 1 per 25 samples), • Certified reference material (one per 25 samples) was inserted by AngloGold Ashanti. • The laboratory prepared a coarse reject duplicate every 25 samples and a pulp duplicate every 20 to 25 samples. • 10 to 20% of the samples were randomly selected and analysed at a second lab (SGS). Standards failure based on a tolerance of two standard deviations was around 3.6% and 1.45% over three standard deviations. Some 99.16% of the coarse blank (sandstone from local quarries) assayed at detection limit or below twice the detection limit of 10 ppb. Only 0.05% coarse blank assayed above 25 ppb Au. The sample batch was accepted if only one standard failed between 2 and 3 standard deviations. The batch did not pass when both standards failed between 2 and 3 standard deviations or when a standard failed over 3 standard deviations. 8.3 Sampling governance The company maintained a site-specific protocol for drilling (Protocolo de Perforacion). The sub-chapters of the protocol are 1.1. Manual drill rig mobilisation, 1.2. Location and orientation of the drill rig, 1.3. Platforms, 1.4. Transport of core boxes to the core shed 1.5. Pre-Logging, 1.6. Core cutting and core sampling, 1.7. Daily drill report, 1.8. Downhole survey, 1.9. Drillhole monument after completing drilling, 1.10 Industrial safety and environment. The QP in his past position as Geology Manager led the compilation of the protocol and carried out the internal QAQC to ensure that no bias or deviation occurred. The drilling platform and trail construction process was also included in the OHSAS 18001 safety certification. The entire drilling equipment was made man transportable. Approximately 3,500 kg of equipment needed to be moved from platform to platform for the mid-sized Kluane 1000 drill rig. The company insured safe drill AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 43 moves, all labour had their respective medical and accident insurance. The project had its own water supply system with water pipelines, central and auxiliary tanks, diesel, gasoline, and electrical pumps. The drill hole orientation was marked using a compass, with a string tied at the ends. The size of the drilling equipment, its actual location on the platform was taken into account. The drill rig was placed on wooden supports which were accommodated according to the azimuth of the drill hole. The inclination of the drill hole was set with the drill inclinometer and checked by the company drill technician. An AngloGold Ashanti drill auxiliary person was onsite for all shifts. The drilling company recovered the drill core from the inner tube and placed it into a tray. The length of the tray depended on the length of the inner tube used by the drilling company. The company auxiliary person was accountable for carefully placing the drill core from the tray into the wooden core boxes. Special instructions were given for broken core, weathered material, or fault gouge to avoid loss of fines or oxides on fracture planes. The broken core was reassembled as much as possible in the core boxes to avoid bias in the later sampling process. The drill core diameter used at La Colosa was dominantly NTW and HTW, to a lesser extent NQ, HQ, PQ, or BTW. Drill core recovery was obtained from the difference between the drill run and the core recovered in the inner tube. An initial value for RQD was obtained on-site. The company drill rig auxiliary person was responsible for correctly marking up the core boxes. This included the ‘From’ and ‘To’ markings shown on the core boxes, also marking the drill depth (meters) on the wooden core box and inserting a wooden separator with the specified drill run, and an additional wooden separator for core loss greater than 10cm. The drill rig auxiliary person also controlled the drilling hours, change in drill rod diameters, difficulties when intersecting disturbed or swelling clay sections, the possibility of the core not recovered/dropped in the drill hole, redrilled drill core, and the correct use of drill additives and muds. Clay and ground rock obtained in the water recirculation system was dried and safely disposed of. Newspaper was used to fill up empty spaces between the drill core and wooden liners, similarly between the drill core and the lid. This, to avoid that the drill core could moving during transport, and also to avoid that broken (potentially high-grade) core could systematically contaminate across the entire core box. The core boxes were strapped up for transport. Core box transport was carried out by local labour using rucksacks (locally fabricated for the size of core boxes), and then with 4 x 4 pick-up vehicles to the core shed, either to the temporary core shed at the La Colosa camp or direct to the El Aceituno core shed. The line of custody was complete. Daily reporting was carried out routinely. It included active drill holes, meters drilled, recovery, and daily issues, a rapid log for lithology, alteration, and sulphide content. The geologist assigned to pre-logging also checked the core boxes for the correct borehole, the ‘From’ and ‘To’ marked on the core boxes and the drill runs. Core cutting and sampling were carried out at the El Aceituno core shed in Ibague. The team consisted of local labour, a person certified for core cutting, and a lead person. The drill core management steps at the El Aceituno core shed end with having the samples ready for dispatch to the commercial laboratory. The initial step after laying out the core boxes and removing the newspaper consisted of inserting red-painted wooden markers every two meters for the respective sample intervals. The red separators indicate on the front the "From-To" of the sample and the separators show on the top the sample number (as used in the database). The task of inserting the sample markers is carried out by the lead person under the supervision of a core logging geologist. Core photos are subsequently taken with both dry and wet core, commonly in sets of two boxes. The drill core is marked with arrows indicating the downhole direction, the fractured core is wrapped with masking tape to prevent the disintegration of the sample during cutting. The core is marked with a line for cutting that must be perpendicular to principal veinlet orientation or foliation. Core cutting was carried out with a brick saw using a 14 diamond-set blade for abrasive material. The project used AngloGold Ashanti sample tickets for core sampling. The in-house sample quality control requires that standards, blanks, and duplicates must be included in all batches of samples. A certified standard, a blank, and duplicates (laboratory) were inserted every 25 samples. Each batch starts with blanks. Monica Uribe (database manager at the Medellin office) was accountable for inserting standards and blanks, and for uploading sampling numbers into the database. The packing and shipment of samples were the responsibility of the lead person. The sampled core is placed into double plastic bags, the sample ticket inserted, and the sample number AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 44 additionally marked on the sample bags. The sampled bags are sealed and lined up to be placed into sacks for shipment to the laboratory. No more than six sample bags were put into one sack. The sacks are numbered and marked with the address of the sample preparation laboratory. The project maintained a sample dispatch format indicating the project, the geologist, type of sample, date, number of samples, address of sample preparation laboratory, logistics person in charge of sample shipping, the transporting company, and the serial dispatch number. Drill core samples were shipped by road using 3 to 5 tonne trucks. The company security department controlled the sample shipment until its arrival at the commercial sample preparation laboratory (commonly ALS Chemex, initially in Bogota, later in Medellin). The governance of sampling, sample preparation, analyses and security, and data verification was positively audited by AngloGold Ashanti’s South American management (2011). The report was prepared as a National Instrument 43-101 Technical Report for AngloGold Ashanti Colombia S.A. Additionally, specific protocols were prepared for core logging of Mineral Resource drill core and geotechnical drill core, orientated core, density measurements, geometallurgical tests, and core scanning. Sample transport to the laboratory was overseen by the company‘s security department. The transport of pulps between preparation and analytical lab was carried out and managed by ALS Custody that uses certified providers. Drill core logging was carried out on paper, digitised and checked for errors, and then uploaded into the Century Database. South American management in their 2011 audit performed a double data input, that consisted of a second digitizing of 15% of the database, specifically of the geological logging, and this was compared to the original digitised data. Only 1.7% showed digitization problems, mostly due to the introduction of data that differs from the original paper. This percentage is within the industry acceptable ranges and assures a reliable database. A script was introduced in 2012 to assure consistency such as alteration minerals logged against dominant and subordinate types of hydrothermal alteration. Additional checks from the database back to the logged core (and interpretations on sections and 3D models) were done to ensure that geological features used to differentiate between early, intermineral, and late-mineral porphyries were correctly done and followed the methodology outlined by Sillitoe (2000). The QP in his past position as Geology Manager reviewed incoming analytical results for possible sample handling errors at the laboratory. This consisted of comparing anomalous high assay results in a low-grade section and anomalous low assay results in a high-grade section with the actual drill core. Duplicates (reject material) were assayed when conspicuous results could not be consolidated. In very minor instances, if there was a sample mix-up at the laboratory or samples were suspected to have been dropped, they were replaced with a different sample from the same interval. The database was then updated with the second, correct assay. Similarly, intersections with anomalous ICP results for metals were reviewed against the drill core and the logging. If required, the drill core was re-logged. Density measurements were manually digitised. Typing errors or problematic core samples were rapidly detected by spurious density results. Dip and azimuth of drill holes were recorded on-site and then manually typed into the DHLoggerTM database. In 2008, the geochemical database operated under Oracle and was transferred in 2011 to the SQL Server administered under the SQL Server Management StudioTM. AngloGold Ashanti Head Office carried out Round Robin tests on an annual basis and no material deficiencies were noted. The last corporate Round Robin report for the La Colosa project is for October 2016. Phil Allen, AngloGold Ashanti chief geochemist during the early exploration phase, carried out laboratory audits for the 2008 conceptual study report. This included the ALS-Chemex Bogota sample preparation facilities, the ALS-Chemex Lima analytical facilities, and the Kappes Cassiday and Associates Reno (Nevada, USA) metallurgical laboratory. Initial metallurgical testing of Colombia exploration projects was carried out at the Kappes Cassiday and Associates laboratory. Phil Allen (2008) considered the facilities as adequate. The Colosa geochemical data is of acceptable quality to complete a Mineral Resource calculation. In the same year, Phil Allen carried out check analyses of ALS-Chemex versus SGS (Lima, Peru). A total of

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 45 743 pulps were sent to SGS Lima Peru for check analysis to be compared with the original ALS-Chemex Lima analysis. Of the 743 check analysis, 27 were a sand-blank and 25 were standards. The standard results all returned values within +/- 2 standard deviations of the certified mean. Of the 27 blanks 16 values were below detection, 10 values returned 5ppb and 6ppb, and 1 blank was unacceptable. Check assays with SGS Lima continued until 2015, the database reports 6,753 check assays. A bias error for the samples checked in 2013 was detected. Re-assays were requested, and all batches were re-assayed (Uribe, 2017). 8.4 Quality Control and Quality Assurance AngloGold Ashanti Colombia has implemented protocols on quality control and quality assurance, QAQC. This consists of regular analysis of reference material such as standards, blanks, duplicates. Coarse blanks (two at the beginning of the hole, then 1 per 25 samples) and certified reference material (one per 25 samples) were inserted by AngloGold Ashanti. The laboratory prepared a coarse reject duplicate every 25 samples and a pulp duplicate every 20 25 samples. Some 10 to 20% of the samples were randomly selected and analysed at a second laboratory (SGS). Certified reference materials (CRM, Au): The AngloGold Ashanti standards protocol specifies the maximum and minimum allowed concentrations for standard values as the certified mean +/- twice the certified standard deviation. Current practice allows CRM to report between 2 and 3 standard deviations if in the sample batch no other CRM with a close mean value reports between 2 and 3 standard deviations. Samples related to a failed standard are identified for possible re-assay when a standard failure has been identified. Identification is based on the gold grade in comparison to the CRM grade, the assay batch, and the location of the sample relative to passed and failed CRMs. The QC data are reviewed periodically in meetings with ALS Chemex and the reanalysis program is agreed on. For coarse blanks, the AngloGold Ashanti standards protocol recommends a maximum limit of twice the detection limit for the respective analytical method. Current practice allows a maximum limit of 5 times the detection limit when a high-grade gold sample was affecting the coarse blank. The database manager is accountable to review the performance of CRMs for each individual batch. The graphs show the analysis of each CRM with respect to the value of certification, the minimum and maximum of 2 and 3 standard deviations. Additionally, a multi-element review of the CRMs is carried out using maximum/minimum graphs with respect to the mean of the historical data in the database. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 46 8.5 Qualified Person's opinion on adequacy The sample preparation and analysis were industry standard. The sample preparation laboratory facilities in Colombia were initially set up in a joint effort between AngloGold Ashanti and ALS Chemex. Specialised personnel from AngloGold Ashanti (Phil Allen) reviewed the laboratory during the time it was in use. Geometallurgical tests for gold recovery proxy data used Leachwell™ analysis and Shake Leach analysis. None of them are included in the Mineral Resource estimation. Gold LeachwellTM analysis proved to be useful for modelling whole ore leach metallurgical recovery tests. The analysis protocol was only partially industry standard as both the sample quantity and leach time had to be optimised by the metallurgical laboratory. LeachwellTM analyses were subsequently carried out by ALS Chemex following the refined protocol. Shake Leach analysis did not prove useful for geometallurgical proxy tests. 9 Data verification 9.1 Data verification procedures The La Colosa project database was managed by Monica Uribe and is a secure, auditable database. All downhole logging data and assay data used for the Mineral Resource estimate were queried from the database. Only topography data, though signed off by the in-house surveyor, was used directly for surface construction (dtm). The geological logging followed a well-defined procedure, originally introduced by AngloGold Ashanti’s greenfield team supported by international consultant R.H. Sillitoe. This led to the distinction between early diorite, intermineral diorite, and late porphyritic intrusions (Sillitoe, 2007). The early and intermineral parts of the stock are hornblende-biotite diorites whereas the late mineral component is hornblende-biotite quartz diorite/dacite; the latter name was preferred in conformity with current project usage. Petrographic studies (thin sections, quantitative XRD, XRF) were carried out to confirm lithologies and mineralogy. A geochemistry study was carried out under a contract with Scott Halley (2012) for additional classification of intrusive rocks and to give insights to the protolith(s) of metamorphic rocks. Cross checks between Mineral Resource logging and geotechnical logging indicated the need to separate volcanic ash, oxide, and transition material from generally all sulphide material. Oxides can contain pockets of sulphide material (soft sulphides) and account for the need to create a greater than 1000 ppm sulphur wireframe overlapping with the oxide and transition wireframes. Volcanic ash displays interlayers of soil and colluvial material and often a paleo-soil layer at the base. It was not possible to correctly sample the gold-bearing colluvial blocks (larger than 20cm) in the ash sequence, as this would have created a bias in the Mineral Resource model. Volcanic ash was treated as a post-mineral event and, subsequently, does not contribute to the Colosa Mineral Resource. 3D geological wireframes, oxide transition wireframes, a sulphur wireframe for soft sulphides, a wireframe for greater than 3% skeletal pyrite, mineralised envelopes, and a topographic digital terrain model (dtm) were prepared following the drilling, logging, and sampling/assaying. The original 3D models were prepared from carrying out geological interpretation on sections. It was evident early on, that gold grades would drop once drilling crossed the lithological contact early diorite/intermineral diorite. Even so, an increase in gold grade was evident in intermineral diorite in proximity to the intermineral diorite/hornfels contact. Sillitoe also introduced the term skeletal pyrite, which was included in the logging protocol. The genetic meaning and relevance to pyrrhotite-alteration was discovered later. The decision to sample drill core at 2m intervals was taken at the start-up of drilling and had not changed throughout the study phases. The geology manager (now QP) managed a team of 4 to 6 geologists, the team received training in logging procedures, core handling and sampling, and, subsequently, in specialised software to create 3D models. The role of the geology manager was to perform guidance and ongoing checks on the advance of the geometry of the geological model and mineralisation. This required comparison between the geological logging, interpretation on sections, and the 3D models. Certain contacts between early diorite and intermineral diorite are not sharp intrusive contacts and required continuous checks between geology and assay. To improve the structural model, a consultant was contracted with the structural geological model underlying the geological 3D model was prepared by Johannes Horner (iC Consulenten). The initial structural geological model of the deposit was introduced in 2011, the last update for the deposit and the infrastructure sites was completed in 2016. J. Horner also reviewed the advance of 3D models in the context of structural AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 47 controls and audited the geotechnical studies. The initial mineralised envelope (ENV05) was prepared in 2008. Envelope ENV05 is based on assay composites averaging 0.5g/t over a minimum width of 10m. Compositing was allowed to include internal waste (irrespective of very low-grade) up to 10m in long intercepts. The resulting drill hole file, named hole05, was plotted on 100m spaced W-E sections. The geology manager/QP carried out an interpretation of mineralisation following the concepts presented by Sillitoe (2000). The interpretation was digitised and the strings linked to obtain an enclosed volume. The resulting mineralised envelope ENV05 was reviewed by Bloy Resource Evaluation for its robustness. Management requested the addition of exploration upside mineralisation at depth due to the initial Mineral Resource estimations and pit optimisations being constrained in depth by the graphic interpretation of the mineralised envelope. The 2017 Mineral Resource estimate is based on three mineralised envelopes: ENV05 (explicit, interpreted on sections), ENV15 and ENV20 (both carried out in LeapfrogTM software). Implicit envelopes had the advantage to better interpret the continuity of mineralisation at certain defined average grades, however, required multiple iterations and reviews to define the amount of internal waste to be included. Top cuts of high-grade assay results were applied constructing implicit envelopes. For control, declustered average grades were calculated in DatamineTM software. 9.2 Limitations on, or failure to conduct verification Data verification was a routine process. 9.3 Qualified Person's opinion on data adequacy The La Colosa project initially completed a conceptual study and obtained board approval in 2008, and a PFS commenced in September 2009. The deposit is considered in social media the “eighth-biggest" undeveloped gold deposit in the world. As such, the deposit has accumulated an extensive technical database on geology and mineralisation, structural geology, geometallurgy, acid mineral drainage and metal leaching, hydrogeology, geotechnical studies, Mineral Resource studies, metallurgy, mining, and infrastructure studies. The data discussed in the Technical Report Summary refers to the 2015-2017 knowledge base. Only sporadic desktop studies for open pit design, infrastructure designs, and metallurgical alternatives occurred during care and maintenance. 10 Mineral processing and metallurgical testing 10.1 Mineral processing / metallurgical testing Both the WOL and the FFG flowsheets make use of common and well-established processes. The tailings filtration circuits will operate at a very large scale and the suitability of industrial equipment for this scale needs to be studied further. As multiple filtration units will be required, the required throughput will be achieved by increasing the number of filtration units. Preliminary sizing of the filtration circuits indicates that the number of filtration units is reasonable. However, with more filtration units the complexity of the plant and layouts will increase in the mountainous terrain. Hence these factors will be an important consideration in determining the optimum plant size. 10.2 Laboratory and results Both the comminution and recovery test work were undertaken by SGS Lakefield, Ontario, Canada from 2011 to 2015. The registrant has no personal or commercial interests in the laboratory. The laboratory is officially recognised by the Standards Council of Canada (SCC) for meeting the requirements of the ISO/IEC 17025 standard pertaining to analytical, mineralogy, and metallurgical tests. The base case flowsheet is WOL as shown on the WOL and FFG flowsheets. The database of cyanidation results comprises a total of over 150 tests for diorite, schist, and oxides. The estimated overall gold recovery is based on the arithmetic averages of the recovery per ore type, weighted by the percentage of each ore type in the plant feed. ICP multi-element analyses have not shown the presence of potentially deleterious elements in significant concentrations. For example, levels of TOC (total organic carbon, 0.1%), cyanide soluble copper (0.003%), arsenic (less than 30g/t), and mercury (less than 0.3g/t) are very low. The potential occurrence of some organic carbon in a very small and limited zone of the orebody will be examined further in future studies. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 48 WOL and FFG flowsheets The mineralisation at La Colosa consists of two principal ore types, diorite and schist. Both are sulphidic. There is a minor amount of oxide material. The ore is amenable to treatment by conventional cyanidation (WOL) and also by flotation (FFG) as discussed above. The WOL flowsheet is a conventional gold processing flowsheet well suited for an ore amenable to cyanidation. The La Colosa ore responds well to cyanidation with high gold recoveries. The proposed flotation processing flowsheet is common for sulphide ore. The La Colosa ore responds well to flotation but flotation tailings still have economically recoverable gold. Therefore, the flotation tailings are leached in addition to the milled concentrate. This flowsheet uses common technology and can achieve a higher gold recovery than WOL albeit at a higher processing cost. From an environmental perspective, the sulphides are concentrated in a smaller mass which has advantages in controlling potential acid generation from tailings. Comminution data are required to determine the size of the milling circuit. A large number of hardness data is available. The milling circuit was sized by Hatch and results were in line with the calculations from the mill suppliers. Additional comminution variability test work was conducted in 2015. These data will be used to model potential variations in plant throughput resulting from ore hardness. This work has not yet been conducted. Preliminary plant layout work undertaken by Hatch in 2014 indicated that a 23Mtpa WOL circuit can be constructed on the El Diamante site but requires a large amount of earthworks. Further plant layout work is required to demonstrate the viability of plant construction on the El Diamante site and for sites within the restrictions imposed by the La Linea tunnel. Additionally, it is necessary to confirm the earthworks costs (AngloGold Ashanti, 2020). This will be an important part of determining the optimum plant throughput. Both the WOL and the FFG flowsheets make use of common and well-established processing operations. The required tailings filtration circuits are large and the suitability of industrial equipment for this scale needs to be studied further. As multiple filtration units will be necessary, the required throughput will be achieved by increasing the number of filtration units. Preliminary sizing of the filtration circuits indicates that the number of filtration units is reasonable. However, with more filtration units the complexity of the plant and layouts will increase in the mountainous terrain. Hence these factors will be an important consideration in determining the optimum plant size.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 49 10.3 Qualified Person's opinion on data adequacy The project data has been presented to company management. Most of the study options included are standard industry practice. The study on RopeCon™ belt conveyors (Doppelmayr Transport Technology) was considered in past (the year 2015) "not conventional industry practice". 11 Mineral Resource estimates 11.1 Reasonable basis for establishing the prospects of economic extraction for Mineral Resource The reasonable and realistic prospects for economic extraction requirement implies that quantity and grade estimates meet certain economic thresholds and that the Mineral Resource is reported at an appropriate cut- off grade that takes into account extraction scenarios and processing recovery. The orebody has also been optimised in a $1,400 shell. A conceptual pit shell was generated in Datamine NPV SchedulerTM software to constrain the Mineral Resource. The estimate of Indicated and Inferred Mineral Resource, contained within the USD $ 1,400/oz pit shell and at a 0.35g/t cut-off, is 28.3Moz @ 0.84 g/t. Metallurgical test work to date indicates that ore can be treated by both whole ore leach and flotation. Processing recoveries and process costs are different for each flotation flowsheet. The economic thresholds also need to consider earth movement costs related to the construction of a mineral processing plant. The early plant layout at El Diamante requires a flat surface for the plant to be constructed on. This would imply slope cuts of more than 100m. Initial studies on tailings site facilities have considered both conventional and filtered tailings deposition methods. The initial engineering designs point to world-class tailings sites (TSF) and waste rock facilities (WRF). ARD and Metal Leaching (ML) mitigation for tailings and waste adds additional costs to the economic model. The construction of TSF and WRF facilities will require large water diversion constructions. Open pit designs used in the alternative assessment desktop review have pointed to the possibility of grade streaming with multiple cut-offs and stockpiling scenarios. The future open pit shell design needs to consider the storage or an application for post-mineral volcanic ash. There are factors related to environmental, permitting, socioeconomic, political, or other relevant factors which could materially affect the Mineral Resource. The main known factors are the Los Nevados Paramo delineation, permitting requirements leading to a positive EIA study, the La Linea tunnel restriction zones, local political opposition, and other stakeholder requirements. The La Colosa Mineral Resource block model from 2015 has been optimised for 23Mt annual ore throughput. The open pit mining scenario has been the primary proposed mining method since the earliest Mineral Resource studies in 2008. The actual pit shell was optimised with an ore mining cost at $2.1/t, a reference mining cost at $2.3/t, a process cost at $8.98/t, with applicable recoveries for whole ore leach (porphyry 82%, hornfelsed schist 83%, oxide 87%), with a General and Administration (G&A) cost of $1.74/t, a gold price of $1,400/oz, a selling cost (which includes royalty) of $1.73/g, a water treatment cost of $0.77/t, an OPEX tailings management facility cost at $1.16/t and an ore transport plus water pumping at $0.13/t. The geotechnical domains and slope angles have been derived from the Call and Nicholas study (2013, 2015). The block model update for 2017 has only minor changes compared to the previous model, as only around 1,100m drilling occurred between 2016. Therefore, it was reasonable to report the 2017 block model contained in the previously optimised pit shell. No additional, Mineral Resource studies have been carried out since 2017. The project site has a provisional exploration camp with accommodation for 150 people, offices, and general storage facilities. There is a fuel station to store 12,000 gallons of diesel fuel, an electrical substation with an installed capacity of 630KVA, and two emergency diesel power plants with a capacity of 200KVA each. The camp successfully operated a wastewater treatment plant with a UV disinfection system. It has capacity is for 700 people on site. There are two water purification plants, the water treatment capacity is 1 litre/second and 0.1 litre/second, respectively. A fair portion of the exploration site has cellular phone coverage. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 50 The project has two access roads, one to the La Colosa campsite, the other to the west of the La Colosa ridge, and serves the infrastructure areas known as Belgíca, El Diamante, and Site 11. Both access roads have a 25t bridge to cross the Bermellon river. The La Colosa camp road is 6km long. The initial bottom section is also used by the local farming community, the higher section is much steeper, includes narrow switchbacks, and is restricted to 4 x 4 vehicles. The El Diamante road is 7.2km long and used by the farming community along the entire stretch. AngloGold Ashanti has been leasing what was originally known as El Diamante farm to the local agricultural community. The purchase of the Romero farm at the end of the El Diamante road is still to be negotiated by the company. The road includes steep narrow turns and should only be used by 4 x4 vehicles. The La Colosa creek and the El Diamante creek can swell rapidly requiring safety measures during heavy rainfall. The El Abuelo track, accessible by narrow track-mounted equipment, extends 2.9km to the west side of the La Colosa ridge. Both principal access roads connect with the Bogota- Cali Buenaventura Panamerican Highway which is currently being upgraded to a dual carriageway. The 8km long La Linea tunnel was inaugurated in 2020. The La Colosa project is connected by a 2-inch (5.08cm) diameter SCH 40 carbon steel water pipeline to receive and pump purchased water for drilling purposes. The pipeline is about 4.5km long and has four defined pumping stations. The bottom pumping station, El Violin at around 2,000m above sea level, receives the purchased water from water trucks at the Pan-American Highway. The Violin station had been equipped with a 30hp 50gpm pumping system. The second pumping station at Antigua Vara at 2,400m above sea level had similar pumping equipment. The third pumping station at El Ordeñadero at 2,600m above sea level was equipped with a 36.4hp 53gpm electric pumping system. The fourth pumping station called ZAC (Central Storage Zone) was designed as the main water storage facility in proximity to the La Colosa camp. ZAC had a water storage capacity of 105m3 and delivered water to platforms required for Mineral Resource drilling. The final point of the water pipeline is called Line 99 at 3,220m above sea level. The distance from El Violin to ZAC is 3,600m and an additional 900m to Line 99. From Line 99 onwards water was then distributed by gravity to the different sites in the entire project area. This included infrastructure drill sites for both hydrogeological and geotechnical drilling at Site 11, El Diamante and Los Pinos. The total capacity of water storage at La Colosa in 2017 was 900m3. The project site registered five drill holes with artisanal water flow, drill hole COL224 registered 73psi (5.03 bar) and 22 litres per second. There is an opportunity to apply for a groundwater water concession. Exploration drilling at La Colosa was carried out with man-transportable equipment. The company built and maintained some 30 km of trails until 2017. This allows groups of more than 100 people to operate the drill moves. An average of 3.5 tonnes of heavy equipment had to be moved between drill platforms along these trails. The project access roads are not conducive for truck-mounted drilling equipment such as RC-rigs mounted on trucks commonly used for grade control pattern drilling or for hydrogeology for pumping tests. Such equipment types would have to be mounted on excavators or Morookas. The access trails in the past were maintained at a width of 1.5m. It is hoped that new permitting will allow wider trails greater than 2m width, especially for principal trails or abandoned old road sections. During mine construction, the entry of machinery and heavy equipment is likely to be from the Port of Buenaventura on the Pacific coast. The port of Buenaventura is some 265 km from the La Colosa project site, and the largest and most important port in Colombia. Alternatively, there are three main ports for heavy equipment to enter via the Atlantic coast. The ports are at the city of Barranquilla, Cartagena, and Santa Marta. The average distance to those ports is around 1000 km. Large cement companies operate close to the La Colosa project site. These are CEMEX in Ibague (Tolima Department), and ARGOS in Medellin (Antioquia Department), and Cali (Cauca Valley Department). The cement companies have the capacity to install concrete production plants in the facilities of the La Colosa site. AngloGold Ashanti has obtained force majeure status on its La Colosa porphyry gold project until June 2022. The main reason has been the inability to advance with exploration due to the lack of environmental permits. AngloGold Ashanti anticipates that all activities required in AngloGold Ashanti’s PFS and FS standards will be approved and will lead to the issuing of an EIA and a PTO document. The economic model could be AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 51 severely curtailed by the Los Nevados Paramo delineation, and community issues could extend the timeline for obtaining a social license required to construct the mine. However, an economic extraction is considered reasonable and realistic from a legal standpoint. La Colosa is a gold operation which intends to sell gold at market rates which in most cases is the spot price at the time of sale. On rare occasions, AngloGold Ashanti may elect to hedge gold sales in advance in which case the price and quantity of the sale is pre-determined according to AngloGold Ashanti’s view of the future gold price and where the price of gold has been contractually set. No marketing parameters are therefore considered for gold sales for the project. AngloGold Ashanti's gold price for the project is set from the company's view of the future gold price environment. This is normally set to a lower value than the actual price expectation, to allow the project to operate at a positive economic margin. The gold price for the project business plan in October 2016 was set at 1,220 $/ounce. The 2017 project was economic at this price. AngloGold Ashanti does however regularly review its gold price assumptions which will be relevant when the project moves out of care and maintenance. In addition to the project gold price, AngloGold Ashanti sets its Mineral Resource and Mineral Reserve gold price on a yearly basis with this based on the company's view of the future gold price inclusive of the need to provide a positive economic margin. For 2017, the Mineral Resource gold price was $1,400/ounce which has since moved to $1,500/ounce. Considering that the project is on care and maintenance, the Mineral Resource has been left as it was determined for the project (at a Mineral Resource gold price of $1,400/ounce), with the view that increase in Mineral Resource gold price continues to adhere to the principle of reasonable and realistic prospects of economic extraction. All costs for the project have been determined through AngloGold Ashanti’s pricing model which is based on the company's view of the project capital as well as future mining and processing costs. All of these were included in the economic model for the project when it was last undertaken in 2017. The La Colosa project is currently at an early project stage and has identified a number of possible technical options all of which are capital intensive. The political risks associated with the mining industry in Colombia, specifically in the Tolima Department, must also be considered. The delineation of the Los Nevados Paramo by Resolution 1987 in November 2016 is considered a risk to the Mineral Resource and it is currently being contested. This puts 13.99Moz of Mineral Resource at risk. The failure to grant environmental permits for site operations has hampered progress and it is the reason that force majeure was accepted by the government. The Mineral Resource is constrained in an optimised pit using a gold price of $1,400/ounce. The QP believes this gives support that the Mineral Resource has a realistic chance of extraction. 11.2 Key assumptions, parameters and methods used The point of reference for this report is 31 December 2021 with all Mineral Resource reported in situ within the optimised Colosa Mineral Resource open pit shell. The data in the report refers to the 2017 Mineral Resource model and the study associated with that. La Colosa does not report a Mineral Reserve, therefore, exclusive and inclusive Mineral Resource are identical. A block model was generated by ordinary kriging performed into a parent block size of 50m (X) by 50m (Y) by 10m (Z) for lithological domains (wireframes) in mineralised envelopes (grade shells). The project maintains lithological wireframes for early diorite, intermineral diorite, late porphyry Intrusions, hornfels, metasediments, San Antonio lithologies, oxide, transition, and post-mineral volcanic ash. Grade shells have been constructed for low grade (ENV05, average grade >0.5g/t), for high porphyry grade (ENV15, average grade around 1.5g/t) and for very high-grade (ENV20, average grade >2g/t). Portions of lithologies outside ENV05 are named waste. The project has defined 25 kriging domains: • Domain 1, Domain 2, Domain 4 Early diorite, intermineral diorite, and hornfels inside ENV05 • Domain 5, Domain 6, Domain 8 Early diorite, intermineral diorite, and hornfels inside ENV15 • Domain 21, Domain 22, Domain 25 Early diorite, intermineral diorite, and hornfels inside ENV20 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 52 • Domain 3, Domain 7 Late porphyry lithologies outside ENV20 • Domain 23, Domain 24 Late porphyry lithologies inside ENV20 • Domain 53, Domain 54, Domain 55, Domain 56 Early diorite, intermineral diorite, hornfels and metasediments outside ENV05 • Domain 31, Domain 32, Domain 33, Domain 34, Domain 35 San Antonio lithologies (no grade shell could be defined for San Antonio domains) • Domain 12, Domain 13 Oxide, transition • Domain 15 Volcanic ash (Au = absent) An empty geological block model was flagged for lithology and mineralised envelopes. A domain identifier was assigned according to the defined kriging domains. Similarly, the drill hole file has been flagged with above defined kriging domains. Density values in the geological block model are based on declustered averages of the measured values for the specific kriging domain. Extreme gold values were evaluated and capped on probability plots. Variography was completed on the 2m (composited) flagged kriging domain samples applying two (or three) structure spherical models. Variograms were calculated for the principal lithologies (hornfels, early and intermineral diorite) in the mineralised envelopes ENV05 and in the mineralised ENV15. An overall variogram was prepared for ENV20, however, then adjusted for declustered variances of the individual lithologies in the mineralised envelope. For the variography of the late porphyry dacite, and dacite dykes only ENV20 was considered as an independent kriging domain, the remaining volume was coded by lithology and did not consider either ENV05 or ENV15. The oxide-transition boreholes were added to the data set for variography. The variogram was then rescaled to the declustered variances of oxide and for transition. Variography for waste borehole domains (outside mineralised envelopes) and the San Antonio lithology boreholes was carried out independently for the specified lithology wireframe and its location. Search ranges for kriging have been estimated with an optimisation script for kriging tests (QKNA script/optimize in Bloy Geostats Toolkit (from Bloy Mineral Resource Estimation) using the actual drill layout of the respective kriging domains. In addition, the maximum number of samples used in the ordinary kriging step was optimised in the same QKNA script. The kriged block model has been visually evaluated for consistency and errors on sections for blocks inside the mineralised envelopes (ie excluding waste, late dacite, and San Antonio). The block model has been validated, initially graphically, then using slice plots; initially section by section (W-E and N-S) comparing block model grades for Au against the Au grade distribution in drill holes. The statistical validation of the overall block model is based on descriptive statistics for block model grade, kriging variance, number of samples per block, slope of regression, and negative kriging weights. The results obtained were validated on sections. Estimated average Au and volume/tonnes were compared to non-estimated volume/tonnes for each kriging domain, and drill hole statistics for gold were compared to block model statistics for gold. For Mineral Resource classification, the distance of blocks to the nearest three drill holes have been added to the kriging output block model. Indicated Mineral Resource was defined in an earlier simulation study and showed a hole spacing less than 75m was required and Inferred Mineral Resource for average drill hole distances greater than75m and/or blocks only estimated by one drill hole). Certain sectors and domains were downgraded for geological reasons. Previous Mineral Resource classifications also validated classification using the two Indicator Method (Tonnage Indicator/Metal Indicator). The latter is based on kriging variance after having estimated the error above cut-off grade with less than 15% error and 90% confidence for a one- year production panel. Results obtained, in general, were similar. Pit shell optimisation was carried out for a gold price of $1,400. The technical input parameters were maintained as defined in the 2014 conceptual study update. For resource reporting, pit shell optimisation only used Mineral Resource gold grades and lower confidence block grades were set to Au=0g/t. The cut-off grade has been obtained using the standard cut-off formula: COG = C / (R * P) where: COG = the cut-off grade expressed in mineral concentration per tonne (g/t), C = all associated costs post the ore/waste decision point, R = mineral recovery expressed as a percentage, and P = effective price expressed as $/g.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 53 The following table outlines the cost key parameters used in the Mineral Resource optimisation. Cost inputs Unit La Colosa Pit Ore Mined (M+I+I) K tonnes 1,051,382 Waste Mined (M+I+I) K tonnes 843,171 Total mined (M+I+I) K tonnes 1,894,554 Stripping Ratio t:t 0.8 Costs Ore Mining cost $/tonne mined 2.1 Waste Mining cost $/tonne mined 2.3 Processing cost $/tonne treated 8.98 G&A $/tonne treated 1.74 Other Parameters Met. Recovery Rock type 1 % 82.2 Rock type 2 % 83.3 Rock type 3 % 87.3 Slope angles Weathered rock and volcanic ash degree 34 High wall degree 48 Schist degree 48 Intrusives degree 50 Mineral Resource Cut Off grade g/t 0.35 Mineral Resource price $/oz 1,400 All geological units have been modelled as 3D solids. The information combines geological outcrop mapping, structural geological mapping, and diamond drilling. As such the wireframes integrate information obtained at different scales. The construction of mineralised envelopes is entirely based on drilling results. The general La Colosa drilling grid was 100m x 100m, including some radial drilling to control mineralisation between 100m sections or to extend mineralisation into upside areas. Initial lithology wireframes (during the conceptual and the initial PFS phase) were constructed using W-E sections on 100m centres and then linked. The final wireframes are based on implicit interpretations carried out in specialised software. This allowed better use of the structural geological controls. The interpretation of high-grade mineralisation and its upside potential appears to be encompassed and truncated by the La Ceja Fault and the La Colosa Fault. The actual upside mineralisation modelled in mineralised envelope ENV15 follows a structural corridor defined by N-S faults, such as NS3A, NS-4, NS-6, and NS-7. The parent cell size for the geological block model is 50m (X) x 50m (Y) x 10m (Z). X and Y dimension is about half of the general drill spacing. The Z dimension is about 5 times the sample length and in accordance with 10m bench heights proposed by the mining department. The dimensions of the block model are sufficiently wide to include a portion of the infrastructure area to the west. Additional solids have been prepared for the infrastructure area to allow infrastructure to be incorporated in the pit optimisations should they need to be. All drilling at the La Colosa project site has been diamond drilling. The open pit area has been covered with 1:1000 total station topographical surveys to link drilling information with surface mapping information and constrain the block model for Mineral Resource reporting. All drill core has been logged, and this was carried out at the 2m sampling intervals. The logging geologists had been thoroughly trained in the protocols for geological logging, geotechnical logging (including Dempers methodology), kenometer logging of orientated core, and in-house soil laboratory methods. Internal audits and reviews, including database management, were in place. Risks AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 54 Ore and waste, with more than 2% sulphur, is potential acid generating, and metal leaching has to be modelled in the site-wide water quality model. Open pit optimisation indicates a greater than 800m high wall at the northern end of the constrained deposit and this will need to be carefully designed and managed. The Los Nevados paramo delineation is the main risk of the project. The delineation (which is being legally disputed) affects Mineral Resource and Mineral Reserve estimations. Also, geotechnical and hydrogeological models become increasingly complex to cater of the exclusive of the paramo. The Mineral Resource block model reflects the year 2017 knowledge base. There are new techniques for creating grade shells and added techniques in the ordinary kriging process. AngloGold Ashanti has been working on a geometallurgy comminution model. Intermineral diorite is harder than early diorite and pit optimisation needs to take those parameters into account. Geometallurgy comminution - tonnes per hour (tph) geometallurgical throughput estimates. In the past years, AngloGold Ashanti carried out initial desktop studies applying localised uniform conditioning (LUC) techniques to the ordinary kriging block model. This allows for much more selective mining, grade engineering, stockpiling, and estimating waste in more detail. LUC benchmarking, especially the definition of selective mining unit (SMU) size for different mining scales, can affect the final Mineral Resource. A change in the overall pit shell angle for geotechnical sectors has a significant impact on the declared Mineral Resource. The La Colosa pit shell displays a highwall higher than 800m for the northern geotechnical sectors. Steepening the overall high wall angle will deepen the optimised pit shell and increase the final Mineral Resource figure. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 55 La Colosa inclusive Mineral Resource grade and tonnage curve 11.3 Mineral Resource generation steps The Mineral Resource generation process followed the steps: • Database construction • Generation of wireframes for lithologies and mineralised envelopes, and surface topography • Geological block model • Grade estimation and localised uniform conditioning? • Generation of Mineral Resource block model • Resource classification • Generation of pit shell • Mineral Resource tabulation Database construction The drill hole file has been queried from the Century System database. 83.4% of the samples were 2m and 16.6% less than 2m (mostly due to the HOLES3D process in Datamine splitting the assay interval for logged lithologies). Therefore, in a later step, the samples were composited for 2m inside respective kriging domains. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 56 Generation of wireframes The geology and the mineralisation at the La Colosa porphyry gold deposit are defined in the block model by using 3D modelled units and surfaces. These are (with domain or file names in brackets): • The topography (dtm17) • Mineralised envelope for greater than 2g/t (ENV20) • Mineralised envelope for greater than 1.5g/t (ENV15) • Mineralised envelope for less than 0.5g/t (ENV05) • Waste/low-grade outside ENV05 (WASTE) • Post-mineral volcanic ash (ash) • Oxide (ox), transition 1 (_trans) • Sulphur less than 1000 ppm (S_1000) • Early Diorite (Wire_E), Intermineral Diorite (Wire_I) • Late porphyry, main dacite body (Wire_Da), late porphyry, dacite dykes (Wire_L) • Hornfels and schists (Wire_H), meta-sediments (Wire_M) • San Antonio, abuela breccia (SA_bxabu), San Antonio, schist breccia (SA_bx) • San Antonio, hydrothermal breccia (SA_hybx), San Antonio, Greystar breccia (SA_hybx2) • San Antonio, late porphyry intrusions (SA_late) • Wireframe for >3% skeletal pyrite (Skpy) Additional lithology wireframes have been prepared for infrastructure site lithologies but do not contribute to the Mineral Resource estimation process. Geological block model The geological block model is the starting point of model building geology and occurs prior to grade estimation. The proto box for the 2017 block model has been maintained from previous block models to accommodate a certain amount of infrastructure condemnation drilling. The parent block size is 50m (X) x 50m (Y) x 10 (Z). The X and Y scale approximate half of the overall drilling grid. Z is 5 times the sampling interval and is also a common bench height for bulk open pit mining operations. Also, SMUs generated for the localized uniform conditioning processes were allowed to fit geometrically into the parent block size and this trialled block sizes of 25m x 25m x 10m, 12.5m x 12.5m x 10m, and 10m x 10m x 10m. The empty geological block model was flagged for lithologies and mineralised envelopes. Grade estimation The gold grades were estimated for the defined kriging domains listed above. Generation of the Mineral Resource block model The output block model is the sum of all kriged domains and the added volcanic ash. The output block model is shown for a porphyry and a hornfels section. The block model has been passed then through several steps of validation: Graphically, section by section (W-E, N-S) comparing block model grade (Au) against Au grade distribution in drill holes; statistical and visual validation of block model grade (Au), kriging variance, number of samples per block, slope of regression, negative kriging weights; volume-tonnes of estimated blocks against volume-tonnes of non-estimated blocks for each kriging domain; and drill hole statistics versus block model statistics and differences. Estimation and modelling techniques. a – lithology wireframes b – mineralised envelopes c – kriging domains for a porphyry section d – kriging domains for a hornfels section e – kriged domains for a porphyry section f – kriged domains for a hornfels section

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 57 Colosa estimation domains AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 58 Mineral Resource classification, the generation of pit shell, and the Mineral Resource tabulation are the final steps. These are discussed further below. The Mineral Resource estimation process (ordinary kriging) was carried out in Datamine Studio 3TM, version 3.24.25.0 (dating 2014). The actual kriging process was run using a custom-built macro requiring kriging parameter input files in .csv format (estimation reference, search reference, variography reference). The kriging parameter files were prepared by the QP. The explicit mineralised envelope ENV05 was prepared in Datamine Studio 3 using conventional wireframe modelling techniques (digitizing, linking sections, and creating solids). The optimisation of kriging parameters (variography, search, number of samples) was carried out with the Blockbuster Geostats Kit (GSK) commercially available in 2014 by BMRE (Bloy Mineral Resource Evaluation). Datamine Downhole Explorer TM, version 3.22.24.0 was initially used for compositing borehole files. Leapfrog TM Mining version 3.1 was used until 2014 to build lithological wireframes and fault surfaces. Leapfrog TM Geo version 4.2 through to version 6.0 was used between 2015 and 2017 to create the 3D lithological wireframes, fault surfaces and fault volumes, geotechnical domains, and the implicit mineralised envelopes. The actual Leapfrog Geo version used to edit and update the geotechnical domain model is 2021.1.3. Datamine NPV Scheduler TM, version 4.24.67.0 was used for pit optimisations. The block model as input file for open pit optimisation was created by ordinary kriging into a parent block size of 50m (X) by 50m (Y) by 10m (Z) for lithological domains (closed wireframes) in the mineralised envelopes. The late dacite porphyry (except for ENV20), the oxide, and the San Antonio domains were estimated as lithology only. The grade estimation was generally carried out using hard boundaries for lithologies in the mineralised envelopes, implying that only data within the domain of interest were used to inform the block. A soft boundary was only included in the kriging macro for domain 25 (hornfels high-grade envelope ENV20), implying data of the additional ENV20 intermineral diorite was considered. A gold value of 0.00g/t was assigned to the (post-mineral, recent) ash domain in the block model. Density values are based on the declustered averages of the specific kriging domain. The kriged block model has been visually evaluated for consistency and errors for blocks inside the mineralised envelopes. The overall block model grade for the mineralised envelope was compared with block model grades obtained in previous years. The small amount of Mineral Resource drilling carried out in 2016 and 2017 aimed to correct ambiguous geometries foreseen in the mineralised envelopes. For 2017, at 0g/t cut-off, the combined mineralised envelopes carry 1,191Mt of mineralised rock. This amounts to 31Moz @ 0.81g/t. ENV15 plus ENV20 at zero cut-off, block model contains 10.7Moz @ 1.54g/t. Applying a cut-off of 1.65g/t, the filtered block model contains 5.2Moz @ 2.18g/t. The 2017 block model also includes newly constructed wireframes for oxide and transition, and in addition, a greater than1000 ppm sulphur envelope for soft sulphides. The oxide-transition model has been improved by using density values and spectral analysis as boundary indicators. As a result, the oxide-transition Mineral Resource increased to approximately 2.5Moz @ 0.43g/t. About 42% of all oxide-transition material has been flagged as transition and 58% as oxide. Finally, all kriging domains plus ash were combined to create the final block model (colosa21). The final block model was then validated, initially graphically, section by section (W-E and N-S) comparing block model gold grades against the gold grade distribution in drill holes. As in previous block models, the gold grades in root zones of ENV05 are spotty for certain sectors and can contribute to spurious poorly correlated high-grade gold blocks. Kriging results in metamorphic rocks reflect the hard boundaries for ENV05. There are isolated higher-grade intercepts in the hornfels waste which do not amount to a minimum 10m composite and kriging smooths these out. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 59 Changes between the final 2015 and the final 2017 block model are minimal. The bulk of PFS Mineral Resource drilling contributing to the wireframe models was completed at the end of 2014. At zero cut-off grade, the 2017 block model (unconstrained by pit optimisation) reports 4,973Mt of ore, this amounts to 54.9Moz @ 0.34g/t. The 2015 block model (there was no update in 2016), reports 4,497Mt of ore, which amounts to 52.4Moz @ 0.36g/t. Summary statistics and histograms have been prepared for gold grade, estimation variance, number of samples contributing to the blocks, slope of regression, and negative kriging weights, and visualised on sections. The block model was validated by visual comparison of drill hole grade versus block model grade. The grade distribution in the high-grade sector is well defined by both drill core assay and interpolated block model grade. The gold comparison for drill core assay and block model grade at the porphyry - hornfels contact zone is consistent for the scale of geological interpretation. Stronger smoothing of high drill core assay grades is evident for mineralised envelopes in hornfels schist. The high-grade assays may significantly contribute to the continuity of mineralised blocks or at the other extreme get smoothed out in low-grade areas. The block model was additionally validated by comparing the average block gold grade-volume/tonnes of estimated blocks and volume/tonnes of non-estimated blocks for the given kriging domains. Hornfels waste and the main late Dacite intrusion contain the highest amount of non-estimated blocks. They form the biggest lithological boundary units of the La Colosa deposit. The principal lithology in the high-grade ENV20 is early Diorite (74%). ENV15 carries 58Mt of unestimated blocks, mostly hornfels, and to a much lesser amount intermineral diorite. It was expected that lithologies inside ENV05 will display unestimated blocks because the geological upside for extending mineralisation at depth was considered in the envelope. The difference in the average gold grade for ENV05 lithologies is narrow (di05: 0.69g/t, idi05: 0.60g/t, hf05: 0.66g/t). The average grade of waste lithology blocks outside the mineralised envelopes is around 0.2g/t. The block model was also validated by comparing mean-maximum-minimum gold grades of borehole assay with interpolated block model grades. This was an iterative process in previous block models, as the capping of high-grade outliers controls the difference of mean grades. The difference was notably visible in kriging domain 8 (hornfels in ENV15, kriging domain 22 (intermineral diorite in ENV20) and kriging domain 25 (hornfels in ENV20). Kriging domain 25 only contains 128 samples, a soft kriging procedure used with kriging domain 22. The San Antonio kriging domains 32, 33, 34, and 35 are lithology only and contain less than 200 samples. The San Antonio porphyry is much smaller and had received less exploration attention. The kriging results were subsequently downgraded from Inferred Mineral Resource. The structurally controlled mineralisation extending north and northwest is the main recognised target to define additional higher-grade mineralisation. The target hosts both porphyry-style mineralisation and late, lower temperature overprints. The latter follows the NS structural corridor, which cuts across all lithologies. Validation using slice plots was the final step before Mineral Resource classification and the generation of pit shells. Slice plots were created using the Geostats Toolkit (software script provided by Bloy Mineral Resource Evaluation). Drilling defined Cu and Mo mineralisation lateral and deeper to the actual reported gold mineralisation. Grades were considered too low to warrant an estimate for co-products. The geological context indicates the potential risk of acid rock drainage (ARD) and metal leaching (ML). The average sulphur content of drill hole samples inside ENV20 is 4.2%, inside ENV15 2.7%, inside ENV05 2%, waste 1.5%, oxide averages 60ppm sulphur for the less than 1000ppm sulphur portion, oxide-transition 1.4% sulphur for the less than 1000ppm (soft sulphide) sulphur envelope, and the skeletal pyrite-rich zone in hornfels averages 1.9% sulphur. Environmental characterisation started in 2010 and continued util 2015. The static characterisation program included: • Acid-Base Accounting (ABA) by the modified Sobek method • Metal scan by Aqua Regia digestion with ICP-MS finish • Metal Scan by four acid digestion with ICP-MS finish • Whole rock chemistry by X-ray fluorescence (XRF) • Quantitative X-ray diffraction (QXRD) with Rietveld refinement AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 60 • Optical petrography • Net Acid Generation (NAG) tests (single addition and sequential addition) • Metal analysis on a composite of the sequential NAG liquors • Grain size distributions An evaluation of the acid potential and the neutralising potential in combination allows for classification of the acid generating potential of an individual sample. Based on guidelines for the definition of potentially acid generating (PAG) rock, except for a small number of samples, all the samples tested fall within the classification of likely to produce acid, also referred to as potentially acid generating, or PAG, with ratios of NP/AP less than 1. Those that do not clearly fall into this classification include a few volcanic ash samples, several hornfels/schist samples, and most of the granodiorite samples from the infrastructure area. These exceptions were driven by the lack of significant sulphide content rather than increased neutralising potential. Because the classification is largely dependent on the amount of sulphide present, the sulphur content itself was used as criteria for the initial classification of waste blocks in the block model. Four waste types were defined for this purpose: • Waste type A = very fast reacting (within months) • Waste type B = acid generating with some lag time (a few years) • Waste type C = potentially acid generating, but low S content • Waste type D = non-acid generating (construction rock) Proposed indicators by which each lithological class could be defined by waste type were provided by Shaw (2012). The criteria set out in the matrix shown below was used by the project team to flag spatial classification of Mineral Resource blocks based on ARD potential. This was completed as a Datamine macro for ARD and assigned to the La Colosa block model. The results clearly show most blocks within the block model have sulphide contents greater than 0.6% (for both the porphyry and the hornfels) and greater than 1% for other units – thereby classifying most blocks as PAG. Sulphur values for the block model have been estimated by ordinary kriging, similarly to ordinary kriging for gold grades. The S less than1000 solid has been estimated separately. In addition, empty blocks after the ordinary kriging have been filled with the 50% probability value of the specified kriging domain. Mineral Resource classification, nonetheless, only applies to the gold estimation, not to the sulphur values in the block model. Preliminary matrix for ARD indicators and criteria for geochemical domain definition 11.4 Mineral Resource classification and uncertainty A Mineral Resource simulation and classification study (Silva and Nunez, 2014) was completed. In 2014, the La Colosa Mineral Resource was modelled with two mineralised envelopes, ENV05 and ENV15. The high- grade mineralised envelope ENV20 was introduced in 2015. The simulated study was performed for 45Mt annual production, for 20Mt annual production, and for a pit Phase 1 at 8 Mt annual production. The underlying criteria were:

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 61 • Measured Mineral Resource: volumes or tonnages of a quarter production with error on grades less than 15% calculated at 90% confidence. • Indicated Mineral Resource: volumes or tonnages of 1-year production with error on grades less than 15% calculated at 90% confidence. • Inferred Mineral Resource: This can be modelled with a reasonable amount of information and geological confidence but cannot reach the criteria of error of less than 15% for the proposed tonnage scales. That means, according to the 15% rule: • Measured Mineral Resource = 90% of the quarterly production is within 15% of the original estimate. • Indicated Mineral Resource = 90% of the annual production is within 15% of the original estimate. • Inferred Mineral Resource = estimates falling outside of the above ranges. For this report, the estimated drilling pattern to comply with the 15% rule has been taken from the smallest scenario: • Measured Mineral Resource: 35m x 35m (Class = 1) • Indicated Mineral Resource: 75m x 75m (Class = 2) • Inferred Mineral Resource: 150m x 150m (Class = 3) The classification has been applied to the La Colosa Mineral Resource model introducing BHFLAG in the ordinary kriging macro. The project does not attempt to define Measured Mineral Resource. This most likely would reflect localities with radial/diagonal drilling from the same platform. Accordingly, the kriging macro defines: • Indicated Mineral Resource: BHFLAG =1 or BHFLAG =2 (average less than 75m from the block to the nearest drill holes). • Inferred Mineral Resource: BHFLAG= 3 or BHFLAG=4 (average distance greater than 75m from the block to the nearest drill holes or the estimate is only carried by one drill hole). • Unestimated blocks and ash have been assigned CLASS=-1, to allow the construction of upside and to assign Au= 0.0g/t if volcanic ash. In 2014 and 2015, the results obtained were classified by the two-indicator method (tonnage Indicator/metal Indicator). The method is outlined in the AngloGold Ashanti guidelines for the reporting of Exploration Results, Mineral Resource, and Ore Reserve (Mineral Resource and Ore Reserve Steering Committee, 2021). The method gave comparable results for the drill pattern introduced in the kriging macro. The Mineral Resource classification results were subsequently audited by Bloy (2014). Bloy established a manual override for the Indicated Mineral Resource classification by examining the isosurfaces in section with a 0.6slope of regression. Single drill line support, areas of poor continuity, and areas with poor slope of regression were highlighted. In general, where the drilling grid opens out to 100m, the quality of estimate drops. Complicating factors at La Colosa include the highly variable grid with localised high concentration of drilling (from restricted platforms) and high-grade domains where the low mining cut-off is inappropriate. Actions taken: All blocks estimated by only one drill hole were reclassified as Inferred Mineral Resource. A manual override for Domain 8 = Inferred Mineral Resource, and for areas with poor geological continuity = Inferred Mineral Resource was created in a Datamine macro. The figure on the following page displays the Mineral Resource classification on sections: (a) W-E Section at 494718N, (b) W-E Section at 494000N, (c) W-E Section at 493500N, (d) W-E Section at 492900N, (e) N-S Section at 445200E. The Mineral Resource estimate update was carried out in 2017. The step-by-step process was established in 2008 and has been improved since. The improvements include uncertainties around data knowledge which could be further evaluated and improved, and software capabilities which also improved during the years: High/low-grade management: AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 62 Capping of high-grades was used and followed a hard capping procedure which removed and substituted high gold values with a lower grade obtained from probability plots. Uncapped high-grades would result in a group of extreme grade blocks driving the pit optimisation. Nonetheless, a better approach would be allowing blocks close to the high-grade drill intercept to use the high-grade drill intercepts in the block model grade estimation. Distant blocks should use only the capped grade. Categorical grade shell modelling: Initially, in 2008, all mineralised grade shells were explicit interpretations carried out by the geologist. Later, in 2012 implicit modelling facilitated better continuation and smoother grade shells. Categorical modelling was presented lately as a new tool. It creates more sophisticated grade shells, with shapes, which can change significantly over small distances. This, potentially, allows more geological detail in the geometry of grade shells. Grade shells for underground scenarios: The actual Mineral Resource estimate is based on an open pit-only scenario. Potential exists to improve the Mineral Resource estimation by including underground scenarios. The preparation of complex, narrow but higher grade, lode-type mineralised envelopes will also guide Mineral Resource drilling requirements not put forward to date. Mineral Resource classification AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 63 Soft-boundary estimation: The actual Mineral Resource estimation process includes boundary analyses for the mineralised envelopes. The extracted drill holes for lithologies inside a specific envelope were compared on QQ plots, mainly to evaluate the potential to combine lithologies. Geologically, porphyry mineralisation consists of many pulses over a short geological timeline, therefore, hydrothermal fluids will extend over the limits of porphyry phases. This can be visualised by porphyry veinlets crossing lithological boundaries. Additionally, the late porphyritic dykes contain fragments of better mineralised early and intermineral diorite, and this resulted in occasional intercepts of economic grade in the late dykes. The Mineral Resource estimation process, in general, used hard boundaries. Nonetheless, soft boundary estimation should be validated in future Mineral Resource models. Grade Estimation applying Dynamic Anisotropy: Dynamic anisotropy in grade estimation allows dynamic search volume orientations, to better search in non- linear domains. Search ellipsoids, in oxide and transition wireframes, would benefit if they could follow the continuity of mineralisation, i.e. the dtm surface or footwall contact. This needs to be investigated. Localized uniform conditioning (LUC): The project has an LUC model but this has not been used to generate a pit shell . The LUC block model which has three possible SMUs, and its use is expected to realise an improvement in the average grade of the reported Mineral Resource. It will facilitate later Mineral Reserve estimations. Geotechnical sectors and overall pit shell angles: Pit shell angles were considered conservative. More aggressive pit shell angles will deepen the conceptual pit shell design and increase the overall tonnage of the reported Mineral Resource. Initial studies have been carried out in-house and applied to the conceptual pit highwall design fora small-scale mining scenario. The geotechnical sectors and the pit shell parameters have not been updated with the 2015 and 2016/2017 geomechanical testwork and the last geotechnical domain model. Geometallurgy The project has a geometallurgical comminution block model. The block model includes kriged A*b and Bond Mill Work proxies and a geometallurgical throughput column. The final conceptual pit shell geometry is expected to change by using this data which quantifies hard and soft ore-types. The classification criteria have been presented earlier. These include geological downgrading for specific lithologies or sectors in the deposit. The optimised conceptual pit shell displays the lithologies, tonnes, ounces, and grades for CLASS=2 (Indicated Mineral Resource), CLASS=3 (Inferred Mineral Resource): Class 2/Indicated Mineral Resource: tons-oz-grade • Early diorite: 230,137,462 tonnes, 7,248,601oz, 0.98g/t • Intermineral diorite: 275,544,007 tonnes, 6,784,034oz, 0.77g/t • Hornfels: 267,740,551 tonnes, 7,584,430oz, 0.88g/t • Late porphyry dykes: 10,534,564 tonnes, 252,886oz, 0.75g/t • Oxide: 26,400,492 tonnes, 769,281oz, 0.91g/t • Transition: 23,134,880 tonnes, 712,190oz, 0.96g/t Blocks above 5 g/t: 749,539 tonnes, 150,732oz, 6.25g/t, maximum gold grade in CLASS 2 = 7.5g/t. This encompasses the presently best-studied portions of the deposit. Class3/Inferred Mineral Resource: tons-oz-grade • Early diorite: 31,011,603 tonnes, 658,064 oz, 0.66g/t • Intermineral diorite: 68,121,548 tonnes, 1,317,148 oz, 0.60g/t • Hornfels: 104,556,027 tonnes, 2696543 oz, 0.80g/t • Late porphyry dykes: 520,783 tonnes, 12,028.02, 0.72g/t AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 64 • Late porphyry dacite intrusion: 3676617 tonnes, 56800 oz, 0.48g/t • Oxide: 6,019,362 tonnes, 153,986 oz, 0.80g/t • Transition: 3,984,788 tonnes, 84205 oz, 0.66g/t Blocks above 5 g/t: 148,230.00 tonnes, 26,829.63 oz, 5.63g/t, maximum gold grade in CLASS3 = 6.84g/t Blocks without lithology, no grade: 247,050 tonnes The late porphyry intrusion in the northern part of the pit shell will contribute to 85Mt of waste with the likelihood to encounter grades above cut-off low. The material will need to be studied for construction material applications. An upgrade from exploration upside to Indicated/Inferred Mineral Resource will also be required for geotechnical purposes (pit high wall designs). The San Antonio sector, which is considered small and with slightly different geometallurgical characteristics will need to be upgraded once infrastructure designs require additional condemnation drilling. There are a few blocks without assigned lithology. This is caused by small gaps between lithologies (software errors during Boolean operations at lithological contacts). There is an additional 77Mt of waste material of late porphyry intrusion in that classification category. Oxide and transition blocks and the late porphyry intrusion classed as exploration upside are located at the northern edge of the pit shell. Similarly, as for Class 4, this portion of the pit will receive attention under pre-stripping and construction material discussions. Upgrades in data knowledge will be required for geotechnical purposes. Volcanic ash pre-stripping: 22,803,433 tonnes Volcanic ash is a post-mineral cover and as such does not contribute to the Mineral Resource study. Nonetheless, volcanic ash is an unconsolidated material and portions have been tested for soil geotechnical parameters. An upgrade in geotechnical knowledge will be required to better understand the risk of landslides during pre-stripping and to propose safe pit angles for the pit optimisation studies for future Mineral Reserve studies. 11.5 Mineral Resource summary Conceptual pit optimisation was carried out for a gold price of $1,400. The applicable royalty was included in the selling cost. AngloGold Ashanti did recommend a gold price of $1,500 for the 2021 Mineral Resource declaration, nonetheless, the report is based on work last updated in 2017 and the Mineral Resource price was left at $1,400. The La Colosa project continues under force majeure and has been on care and maintenance since April 2017. The pit optimisation process in 2015 used the Indicated and Inferred Mineral Resource. The table below summarizes the input parameters such as scale of ore production, mining and process costs, gold recoveries, geotechnical domains, NPV discount rate, and other costs. The cut-off grade has been obtained using the standard cut-off grade formula: COG = C / (R * P) where: COG = the cut-off grade expressed in mineral concentration per tonne (e.g. g/t) C = all associated costs post the ore / waste decision point R = mineral recovery expressed as a percentage P = effective price (revenue less revenue related costs) expressed as $/g 12.59/ (0.829 x 43.28) = 0.35 (g/t).

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 65 Pit optimisation input parameters The La Colosa Mineral Resource is reported at a 0.35 g Au/t cut-off. There is no Mineral Reserve at La Colosa and therefore all Mineral Resource is exclusive Mineral Resource. Exclusive gold Mineral Resource La Colosa Tonnes Grade Contained gold as at 31 December 2021 Category million g/t tonnes Moz Measured - - - - Indicated 833.49 0.87 726.31 23.35 Measured & Indicated 833.49 0.87 726.31 23.35 Inferred 217.89 0.71 154.86 4.98 AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 66 11.6 Qualified Person's opinion While the project is on hold, alternate supporting studies are being conducted which will enhance any future options. These studies include any new technologies that may bring value to the project as well as environmental, social or governance (ESG) aspects that will facilitate the future project. Critical economic factors applicable are the gold price and earth movement costs. 12 Mineral Reserve estimates The La Colosa project does not report a Mineral Reserve and accordingly all sections associated with the Mineral Reserve are excluded from this report. 13 Mining Methods The La Colosa project does not report a Mineral Reserve and accordingly all sections associated with the Mineral Reserve are excluded from this report. 14 Processing and recovery methods The La Colosa project does not report a Mineral Reserve and accordingly all sections associated with the Mineral Reserve are excluded from this report. 15 Infrastructure The company has purchased the surface rights or acquired an agreement to access the ground required for exploration activities. The project team studied infrastructure localities for the 23Mt per year case (670Mt ore, 452Mt waste, 30 years life of mine) up to 2017. The studies point to plant site localities, conventional tailings deposition, the construction of freshwater ponds, a contact water pond, and water management structures. The infrastructure studies were accompanied by geotechnical and hydrogeological drilling, All infrastructure localities occur to the west of the La Colosa Mineral Resource in the La Guala, El Diamante, and La Arenosa basins (Figure under hydrogeological and geotechnical studies). There are three potential areas for plant sites: a) at the El Diamante ridge (between the El Diamante and the La Arenosa creek), b) at Los Pinos (inside the Smurfit farm area), and c) Alto Girardot (south of the La Guala creek). Site 11, at the confluence between the La Guala and the El Diamante, was studied for conventional tailings deposition. It includes a tailings/waste co-disposal scenario with a 12.7 Mm3 starter wall and a final dam height close to 400m. The contact water pond, downstream of the co-disposal site at the La Guala creek La Arenosa Creek confluence, was estimated to have a capacity between 4 - 10 Mm3. The location of the freshwater pond was proposed downstream of the contact water pond. Early designs of water management structures included water diversion channels along the northern side of the impoundment, a sidehill structure that would connect to a decant tunnel constructed through the east side of the impoundment and further into an inlet structure. Condemnation drilling confirmed the viability of Site 11 for infrastructure purposes and the ARD investigation and mitigation study is pending. The study does not consider logistics. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 67 16 Market studies The primary product sold from the mining and beneficiation of ore at our operations, is gold doré. The accepted framework governing the sale or purchase of gold, is conformance to the loco London standard. Only gold that meets the LBMA’s Good Delivery standard is acceptable in the settlement of a loco London contract. In the loco London market, gold is traded directly between two parties without the involvement of an exchange, and so the system relies on strict specifications for fine ounce weight, purity and physical appearance. For a bar to meet the LBMA Good Delivery standard, the following specifications must be met as a minimum: • Weight: 350 fine troy ounces (min) – 430 fine troy ounces (max) • Purity / Fineness: Minimum fineness of 995.0 parts per thousand fine gold • Appearance: Bars must be of good appearance not displaying any defects, irregularities such as cavities, holes or blisters. Only bullion produced by refiners whose practices and bars meet the stringent standards of the LBMA’s Good Delivery List can be traded on the London market. Such a refiner is then an LBMA Accredited Refiner and must continue to meet and uphold these standards in order for its bars to be traded in the London market. Provided the bullion meets the LBMA Good Delivery standard, it is accepted by all market participants and thus provides a ready market for the sale or purchase of bullion. Annually, the gold prices used for determining Mineral Resource and Mineral Reserve are determined by the Mineral Resource and Ore Reserve committee (RRSC). Two different prices used for determining Mineral Resource and Mineral Reserve. These prices are provided in local currencies and are calculated using the historic relationships between the USD gold price and the local currency gold price. The Mineral Resource price reflects the company’s upside view of the gold price and at the same time ensures that the Mineral Resource defined will meet the reasonable prospects for economic extraction requirement. Typically, the price is set closer to spot than the Mineral Reserve price and is designed to highlight any Mineral Resource that is likely to be mined should the gold price move above its current range. A margin is maintained between the Mineral Resource and ruling spot price and this implies that Mineral Resource is economic at current prices but that it does not contribute sufficient margin to be in the current plans. The Mineral Reserve price provided is the base price used for mine planning. AngloGold Ashanti selects a conservative Mineral Reserve price relative to its peers. This is done to fit into the strategy to include a margin in the mine planning process. The company uses a set of economic parameters to value its assets and Business plan, these economic parameters are set on a more regular basis and reflect the industry consensus for the next five years. These are generally higher than the Mineral Reserve price and enable more accurate short term financial planning. Finally, the company uses a fixed price to evaluate its project and set its hurdle rate. This price and the hurdle rate are set by the board and changed when indicated due to significant changes in the price of gold. The determination of the Mineral Resource and Mineral Reserve prices are not based on a fixed average, but rather an informed decision made by looking at the trends in gold price. The gold prices and exchange rates determined are then presented to the RRSC for review, in the form of an economic assumptions proposal document once a year (generally the second quarter of the year). After review and approval by the committee, it is sent to AGA’s Executive Committee ("EXCO") for approval. The prices for copper, silver and molybdenum are determined using the same process used for gold. No material contracts required for development have been drafted or issued at this time, given the early stage of property assessment. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 68 17 Environmental studies, permitting plans, negotiations, or agreements with local individuals or groups 17.1 Permitting Before entering care and maintenance in 2017, AngloGold Ashanti Colombia carried out environmental and social baseline studies, and obtained ISO 14001:2004 certification. Exploration followed environmental guidelines, and a plan based on these guidelines was presented to and monitored by the regional environmental authorities. The La Colosa Project (LQP) area is in a forest reserve and as such the company´s project team received two principal environmental permits (temporary forest extraction). The third and last application detailing the exploration needs under EIA studies was denied. The first and second permitting resolutions allowed the construction of drill platforms on grassland and forest openings, the construction of the exploration camp, the associated storage facilities, and an electrical substation, allowed the construction and maintenance of trails leading to drill platforms, and temporarily subtracted the main access roads to project. The ministry resolution requested a detailed hydrometeorological, flora, and fauna monitoring campaign. The monitoring was subsequently carried out by the National University (Medellín) and Conservation International. The project had operated in the past with water concessions and water discharge permits for camp facilities. With force majeure, all exploration and monitoring activities ceased. The regional environmental authority, Cortolima, terminated the water permits. Prior to entering care and maintenance in 2017, the La Colosa project complied with the necessary requirements defined for: • Environmental permits (water concessions and discharge) • Exploration environmental guidelines • Temporary forest clearing • ISO 14001:2004 certification These requirements were presented in reports to the regional and national Environmental Authority and were part of follow-up audits required by the ISO 14001:2004 certification. All the requirements were fulfilled. Permits normally required for exploration activities are: • Environmental permits for water concessions, water discharge, occupation of water channels (creeks), and the use of forest (if necessary). • Temporary forest clearing. The permits required for mine construction and mine operation are: • Environmental License (EIA) • Construction and Mining Plan (PTO) • Definitive forest clearing. The project will revert to an exploration project at Year 1 once force majeure ends and will continue its first three years exploration phase. The company will submit a report to the National Mining Agency detailing all surface and subsurface exploration it plans to carry out. The report includes schedules and investments. There is an environmental mining guide that the exploration activities must adhere to. An environmental impact assessment will be carried out for surface and sub-surface exploration. The environmental management plan will be prepared and included in the platform permitting process. The company has been working on permitting/temporary forest clearing strategy, which embraces a timespan through to FS and include the submission of an EIA report. As infrastructure designs are at pre-conceptual understanding, most permitting tasks require flexibility in relation to the actual location in the field. There is an active ongoing permitting request which was filed with the environmental ministry in 2018.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 69 AngloGold Ashanti believes all permits can be achieved, but that the required timeline can only be estimated at a very high level. The two main sensitive areas that may affect the project are: • The Los Nevados Paramo delineation; and • The restriction zone around the La Linea tunnel. Based on this, alternative mining scenarios must be studied. A study and permitting timeline will be created to detail this for the EIA of the selected mining alternative. Mitigation measures: • For the Los Nevados Paramo delineation, the framework has been described under the Legal chapter. • For the La Linea tunnel, the project can obtain permission from the highway tunnel operator to place infrastructure within the restricted zone. • A detailed plan has yet to be developed. 17.2 Requirements and plans for waste tailings disposal, site monitoring and water management Prior to entering care and maintenance, the La Colosa project operated at the PFS level for most topics. The conceptual study reports were updated for the inclusion of new technology or significant changes in design criteria. While this also applies to the recent desktop studies, a conceptual study phase will need to be conducted to return to PFS when the company re-starts advanced exploration. The waste, tailings disposal, site monitoring, and water management for a future mining operation and the mine closure will be presented to external auditors in the PFS report according to company standards. For the past exploration phase, the waste and site monitoring, and the water management for exploration activities was approved and reviewed by the regional and national environmental authorities. 17.3 Socio-economic impacts A Social Management Plan (PGS) must be presented to the National Mining Agency together with the PTO. In the past, the La Colosa project developed programs for community engagement, community participation, and community-building programs. Previous programs were voluntary. Currently, the La Colosa project is under force majeure, the PGS-related tasks will proceed once the project progresses. During force majeure, there are no socio-economic and cultural impacts that need to be mitigated however the company has studied strategies to quantify and mitigate socio-economic impacts for future projects. The initial study results were presented in a 2021 framing workshop and laid out the chapters for a new conceptual study phase. 17.4 Mine closure and reclamation Mine closure and reclamation plans were detailed in previous studies but in force majeure, they are not applicable. Once force majeure ends, the project will most likely enter a new study phase, and the identification of socio-economic and cultural impacts will once again be undertaken. 17.5 Qualified Person's opinion on adequacy of current plans Issues related to environmental compliance, permitting, and local individuals or groups are examined and developed out by experienced personnel (personnel who worked on the environmental license granted to AngloGold Ashanti's Gramalote gold project and who are currently working at Quebradona copper-gold project). The team has the necessary skills and experience to develop the tasks and budgets required. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 70 17.6 Commitments to ensure local procurement and hiring During force majeure, only local infrastructure care and maintenance activities are carried out at the project site. 18 Capital and operating costs 18.1 Capital and operating costs The present study does not report detailed economic criteria. The La Colosa project is currently at an early stage and has identified a number of possible technical options all of which are capital intensive. 18.2 Risk assessment The political risks associated with the mining industry in Colombia, particularly in the Tolima Department, must also be considered. The delineation of the Los Nevados Paramo by Resolution 1987 is considered a risk to the Mineral Resource and is currently being contested. This puts 13.99Moz of Mineral Resource at risk. The failure to grant environmental permits for site operations has hampered progress and is the reason that force majeure was accepted by the government. AngloGold Ashanti Colombia has decided to put the La Colosa project on care and maintenance and obtained a force majeure status from the National Mining Agency. The project is not expected to be reactivated until satisfactory environmental platform permitting has been achieved. 19 Economic analysis 19.1 Key assumptions, parameters and methods The project does not report detailed material assumptions. 19.2 Results of economic analysis The net cash flow after taxes estimated for La Colosa in the Concept Study completed in 2014 is positive. The mine plan has a mine life of 15 production years plus 2 years of pre-stripping. The strip ratio to expose ore is on average 0.52 for the life of mine, and the total production schedule was estimated at 9.3Moz. IRR and cash flow estimated during the Concept Study completed in 2014 are positive. The payback period is after year 7 of production. 19.3 Sensitivity analysis The project does not report a detailed sensitivity analysis. 20 Adjacent properties In the past, AngloGold Ashanti Colombia greenfields grouped the tenement blocks surrounding La Colosa under Santa Maria Brownfields targets. Montecristo, Tierradentro, Bolivar and La Empresa are similar gold porphyries, but far less studied exploration targets. Age dating of regional porphyry intrusions and mineralisation confirmed a well-defined cluster of porphyry-style mineralisation. The magma chamber accountable for La Colosa most likely extended for greater than 15km. In the process of La Colosa entering care and maintenance, AngloGold Ashanti Colombia decided to ask the National Mining Agency to relinquish the Santa Maria Brownfields targets. Tierradentro and La Empresa are still controlled by the company and involved in a lawsuit with local authorities. No exploration data has been publicly disclosed for the adjacent properties. Information on mineralisation at Montecristo and Bolivar is held internally with AngloGold Ashanti. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 71 AngloGold Ashanti Colombia has carried out prospection and non-invasive exploration work in the adjacent areas. These areas indicate the potential of a Mining District. Regionally, additional potential for Mineral Resource were identified at Montecristo, Tierradentro, Bolivar and La Empresa. The report does not include information from adjacent properties. 21 Other relevant data and information 21.1 Inclusive Mineral Resource At La Colosa there is no Mineral Reserve and therefore the inclusive and exclusive Mineral Resource is the same. The knowledge base for this is 2017 21.2 Inclusive Mineral Resource by-products No by-products are reported for the property. 21.3 Mineral Reserve by-products No by-products are reported for the property. 21.4 Inferred Mineral Resource in annual Mineral Reserve design A Mineral Reserve has not been estimated for the La Colosa project therefore this is not applicable. 21.5 Additional relevant information On March 26th, 2017, residents of Cajamarca’s municipality voted in a referendum to disapprove mining projects in town, including La Colosa. The Mining minister of Colombia subsequently publicly confirmed that Cajamarca’s vote does not apply retroactively, so it does not impact the La Colosa project. On April 27th, 2017, AngloGold Ashanti Colombia suspended all exploration activities at La Colosa project until certainty about mining activities in Colombia are clarified. On October 11th, 2018, the Colombian Constitutional Court issued ruling SU-095 -2018, stating that local municipalities or regions are not entitled to veto mining activities through popular consultations (referendums) on their territories. The Constitutional Court also ordered Colombian Congress (legislative body), to enact a law within two years to ensure that local communities and groups are consulted in the approval of mining activities in accordance with specific criteria set out in its ruling. 21.6 Certificate of Qualified Person(s) As the author of the report entitled La Colosa: Technical Report Summary, I hereby state: 1. My name is Rudolf Jahoda. I am the Qualified Person for the Mineral Resource 2. I am a Specialist Contractor, contracted directly to AngloGold Ashanti, for exploration and mining geology 3. I am a member of the AusIMM (Member of the Australasian Institute of Mining and Metallurgy, registration number 990544) and I have a PhD (Geology) and MSc (Mining Geology) 4. I have 28 years of relevant experience. 5. I am a Qualified Person as defined in the SEC S-K 1300 Rule. 6. I am not aware of any material fact or material change with respect to the subject matter of the report that is not reflected in the report, the omission of which would make the report misleading. 7. I declare that this report appropriately reflects my view. 8. I am not independent of AngloGold Ashanti Ltd as I am contracted directly to them. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 72 9. I have read and understand the SEC S-K 1300 Rule for Modernisation of Property Disclosures for Mining Registrants. I am clearly satisfied that I can face my peers and demonstrate competence for the deposit. 10. I am not an employee of the issuer, AngloGold Ashanti Ltd for the 2021 final Mineral Resource 11. At the effective date of the report, to the best of my knowledge, information and belief, the report contains all scientific and technical information that is required to be disclosed to make the report not misleading. 22 Interpretation and conclusions The main risk to the Mineral Resource is the Los Nevados Paramo delineation and, consequently, to the economically extractable portion as a Mineral Reserve. Failure to achieve a positive outcome in the lawsuit will result in a smaller Mineral Resource/Mineral Reserve, and consequently, a less attractive financial model. The hydrogeological and geotechnical scopes, in particular for a pit highwall design against a Paramo boundary would add to the complexity of environmental permitting. The Alternative Assessments desktop review proposes a smaller project scenario with stringent environmental restrictions. These are: • The pit shell selection is aimed at achieving higher initial feed grades, and minimising strip ratios to increase early revenue. Interim pushbacks will be utilised, with between 3 and 4 pit phases likely. Higher grade mineralised envelopes/zones that have been identified may offer some upside potential for underground operations. • Standard open pit mining equipment is likely to be utilised with conventional drilling, blasting, loading and hauling. Preliminary access to the excavation area may need to use a smaller fleet until the bench space has been opened enough to use the larger fleet. The primary load and haul fleet is expected to be made up of 400t class shovels, and 140t trucks, although this will be subject to review in the next phase of study. The load and haul fleet will be supported by ancillary equipment such as production/grade control/pre-split drill rigs, wheeled loaders, dozers, graders, water trucks, etc. • To maximise value to the operation, grade streaming will be utilised, particularly during the early years of operation. This will result in additional ore, over and above the processing throughput being mined, with the higher-grade ore being preferentially fed. Extra ore will be stockpiled, however due to the terrain around the operation stockpiling space will be at a premium, and it is currently assumed a maximum of 25Mt can be stockpiled at any one time. • Initial waste rock will be used for construction purposes were possible and depending on the final processing route the majority of the waste rock will either be sent to a dedicated waste rock facility or will be co-disposed with the filtered tailings. 23 Recommendations The QP recommends additional work to be carried out for: Paramo studies: The Paramo study should be resumed with an emphasis on elevating the quality of the present Paramo study from standard practice to best (international, PhD to Post Doc) practice. The study should consider a time frame of at least three years. An international university experienced on ecosystem processes and ecosystem services should be contracted to define the details of the scope. The cost of the study should be in alignment with internationally accepted costs of PhD or Post Doc investigations. The actual risk of the Paramo lawsuit is a lack of preparedness for latest international ecosystem methodologies and peer-reviewed results when the lawsuit process progresses. Update Mineral Resource: With moderate effort, the registrant of the report (AngloGold Ashanti) is in the position to update the Mineral Resource statement carrying out a new pit optimisation study. The international Mineral Resource evaluation team has completed a LUC study with selective mining units of 10m x 10m x 10m, 12.5m x 12.5m x 10m, and 25m x 25m x 10m (Silva, A., and Oliveira, A., 2021). The extended, integrated Mineral Resource block model also accounts for geometallurgical comminution proxies which should be included in the pit optimisation process. The present definition of geotechnical sectors should be reviewed in-house before they are used in the pit shell optimisation. The Mineral Resource pit shell optimisation update is to be carried out in-house.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 73 Hydrology database: The climate, creek flow, precipitation, and water quality database were not correctly archived when the project entered care and maintenance in 2017. The hydrology database should be reviewed by local experts after the necessary software updates are implemented. ARD database: The study chapter on ARD has not been compiled and written up. The QP has spent considerable time recompiling the information and the information as such appears to be complete. The ARD database should be audited (internally) for the completeness of laboratory certificates. El Aceituno core shed: The El Aceituno core shed has been shut down for close to five years and degradation of wooden core boxes has occurred. AngloGold Ashanti will need to prepare a budget to replace degraded core boxes. The task is preferably carried out over a period of six months. La Colosa project site: There has not been any field activity at the project site for many years. Secondary forest growth appears to have replaced a fair percentage of the original grass lands. The status of secondary forest growth together with the condition of access roads, the actual camp site, and the physical condition of instrumentation such as piezometers and weather stations should be documented on a yearly basis. 24 References 24.1 References Allen, P., 2008, Colosa QAQC Conceptual Study, Memorandum, 93 pp. Allen, P., 2008, Colosa Check Au Analysis ALS vs SGS Lima Peru, Memorandum, 2 pp. AngloGold Ashanti, 2008, Greenfields Exploration, Advanced Projects, La Colosa Project Colombia, Conceptual Study, 172 pp. AngloGold Ashanti, 2014, 1.0 Protocolo de perforacion, 14 pp. AngloGold Ashanti, 2014, La Colosa On Mountain Option Concept Study Update: Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Concept. AngloGold Ashanti, 2014, La Colosa On Mountain Concept Study, AngloGold Ashanti Group Planning and Technical, Independent Competent Review, 63 pp. AngloGold Ashanti, 2018, Colosa Project, Small Case, Desktop Study, 18 pp. AngloGold Ashanti, 2020, La Colosa Project (LCP), Conceptual Study, Process Engineering, December 2020, RFP, 18 pp. AngloGold Ashanti, 2020, La Colosa Conceptual Study, Request for Proposal for the design of the Tailings Storage Facility, Waste Rock Facility and Associated Infrastructure, December 2020, RFP, 22p. AngloGold Ashanti, 2021, Colosa Project 2.0, Alternatives Assessment, Desktop Review, 47 pp. Arce Geofisicos, 2011, Induced Polarization Survey, February/April 2010, September 2010/January 2011, Report 856B-11, 10 pp. Arce Geofisicos, 2011, Seismic refraction survey with eikonal tomographic interpretation, Survey 940-11, Survey 940A-11, Survey 940B-11, reports, maps and plates. Bloy Resource Evaluation, 2013, Colosa Geomet October 2013, Technical Note, 34 pp. Bloy Resource Evaluation, 2014, Colosa Hard Kriging Boundaries, Technical Memo, 5 pp. Bloy Resource Evaluation, 2014, Comments on Colosa Resource Classification, Technical Memo, 9 pp. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 74 Call and Nicholas, 2013, La Colosa Pre-Feasibility, Geotechnical Report, 165 pp. Call and Nicholas, 2015, La Colosa ISA design update, 56 pp. Castro, A., Drews, U., Jahoda, R., 2015, La Colosa Project, Metallurgical testwork phases 3 and 4, sample selection, 23 pp. Castro, A., Drews, U., Jahoda, R., 2015, La Colosa Project, geometallurgical comminution sample selection, 16 pp. Castro, A., Drews, U., Jahoda, R., 2016, Metallurgical Testwork, pyrrhotite removal from concentrate, sample selection, 19 pp. Dempers, G.D., 1994, Optimal Usage of Exploration Core for Geotechnical Purposes. In: Integral Approach to Applied Rock Mechanics. The 1994 ISRM International Symposium, Santiago, Chile, May 1994, 1, p. 219 230. Dempers, G.D., 2010, Optimising Geotechnical Logging to Accurately Represent the Geotechnical Environment: Second Australasian Ground Control in Mining Conference, Sydney, NSW, 23 24 November 2010, 85 94. Drews, U., Jahoda, R., Leichliter, S., Lozano, C., Montoya, P., 2011, Metallurgical testwork sample selection, 88 pp. Drews, U., Jahoda, R., Lozano, C., Montoya, P., 2013, Phase 2 Metallurgical sample selection, comminution and recovery variability testwork, 36 pp. Geospectral Imaging, 2015, La Colosa Project 2015, Hyperspectral Data Acquisition, Documentation for the sisuMobi Core Imaging Campaign, Instrument Configuration in April-May 2015, 18 pp. Geospectral Imaging, 2015, Interpretation of spectral data from selected samples at La Colosa Project, Mineralogical interpretation of cut core samples from the La Colosa Project (LC2015), 11 pp. Golder Associates, 2015, Balance de Aguas Linea Base, Estudios de prefactibilidad del TMF y WRF, el reservorio de aguas de contacto (CWAP) y obras relacionadas del proyecto La Colosa, 96 pp. Golder Associates, 2017, Reporte factual campana geotecnia e hidrogeologico ano 2015 y 2016, aseguramiento de calidad y control de calidad (QA/QC) de campana de perforacion geotecnica e hidrogeologica, 85 pp. Golder Associates, 2017, Caracterizacion hidrogeologica conceptual detallada, proyecto La Colosa, Sitio 11, 206 pp. Halley, S. 2012, La Colosa Geochemistry, Orefind, 54 pp. iC Consulenten, 2011, Structural Geological Study (PFS) Final Report, 60 pp. iC Consulenten, 2013, Borrow Materials On Mountain Infrastructure Areas, Technical Note-N. 12, 17 pp. iC Consulenten, 2016, Structural Geological Model, La Colosa Project Update December 2016, 44 pp., 11 annexes, Leapfrog model. Ingenieria y Georiesgos, 2013, Analisis de amenaza sismica/analisis de riesgo/impacto sobre la infraestructura, diseno y construccion del pit, 160 pp. Jahoda, R., 2017, PFS Light, July 2017, La Colosa Project, On Mountain Option, 89 pp. Jahoda, R., 2017, Mineral Resources and PFS Planning Report 2017, La Colosa Mineral Resource Statement, 102 pp. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 75 LC-Arcadis -2014- Campana de perforacion Sitio 11, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes memo drill proposal, soil logging style). LC-Golder-2016- Caracterizacion geotecnica Site 11, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes working reports, memos, tests, discussions). LC-Golder-2016- Infrastructure Drilling Campaign, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes the factual report, memos, tests, appendices). LC-Golder-2017- Caracterizacion geotecnica Site 11, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes various reports including Geophysics May 2017, Packer tests, special laboratory tests, analyses, and technical memos). LC- Hidroingenieria-2016 Site 11 Tomography, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes Phase 1, Phase 2 data, images, and reports). LC- Ingetec -2016 Caracterizacion geotecnica Site 11, Folder in AngloGold Ashanti Colombia Sharepoint Site for Colosa/Geology/Geotechnical Studies. (The folder includes Geoprobe tests, GINT logs, laboratory tests, photos, HSE). Mineral Resource and Ore Reserve Steering Committee, 2021, AngloGold Ashanti, Guidelines for the reporting of Exploration Results, Mineral Resource and Ore Reserve, 86 pp. Naranjo, A., 2015, Programa unico de exploracion para la integracion de areas de contratos de concesion EIG-167, EIG-163, EIG-166, GGF-151, GLN-09261X y HEB-169, proyecto La Colosa, 77 pp. Naranjo, A., Horner, J., Jahoda, R., Diamond, L.W., Castro, A., Uribe, A., Perez, C., Paz, H., Mejia, C., Weil, J., 2018, La Colosa Au Porphyry Deposit, Colombia: Mineralization Styles, Structural Controls, and Age Constraints: ECONOMIC GEOLOGY, v. 113, 3, p. 553-579. Nugus, M., Naranjo, A., Horner, J., Castro, A., Uribe, A., and Montoya P., 2014, LaColosa: Project Report (TA: Sam-1-14), 43 pp. Nugus, M., 2016, Review of Geochemical Characteristics and Signatures for the LaColosa, Western High- Grade Gold Domain, AngloGold Ashanti Strategic Technical Group, 22 pp. Nugus, M., 2021, LaColosa High-Grade Domain Wireframes Context and Assumptions, Memorandum, AngloGold Ashanti Group Planning and Technical, Strategic Technical Group, 7 pp. QG, 2012, Mineral Resource Review, La Colosa, Colombia, 87 pp. Seery, J., 2017, La Colosa Highwall Slope Conceptual Geotechnical Parameters, AngloGold Ashanti internal memo, 68 pp. Shaw, S., 2012, Matrix for geochemical domain definition in the La Colosa Block Model, Memorandum pHase Geochemistry, 7 pp. Sillitoe, R.H., 2000, Gold-rich porphyry deposits: Descriptive and genetic models and their role in exploration and discovery: Reviews in Economic Geology, v. 13, p. 315-345. Sillitoe, R.H., 2007, Preliminary geological model for the Colosa Porphyry gold system, Colombia. A report prepared for Sociedad Kedahda S.A., 13 pp. Silva, A., Nunez, L., 2014, Geostatistical Simulation Study for La Colosa PFS Project, Note for the Record, 16pp. Silva, A., Oliveira, A., 2021, Colosa LUC study 2021, Note for the Record, 9 pp. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 76 South American Management, 2011, Audit database of geology, the Colosa project, 31 pp. Uribe, M., 2017, Analysis, La Colosa Conceptual Report, June 2017, 19 pp. U.S. Department of the Interior, Bureau of Reclamation, 1990, Earth Manual, Part 2, A Water Resources Technical Publication, Third Edition, p. 1255. 24.2 Mining terms All injury frequency rate: The total number of injuries and fatalities that occurs per million hours worked. By-products: Any potentially economic or saleable products that emanate from the core process of producing gold or copper, including silver, molybdenum and sulphuric acid. Carbon-in-leach (CIL): Gold is leached from a slurry of ore where cyanide and carbon granules are added to the same agitated tanks. The gold loaded carbon granules are separated from the slurry and treated in an elution circuit to remove the gold. Carbon-in-pulp (CIP): Gold is leached conventionally from a slurry of ore with cyanide in agitated tanks. The leached slurry then passes into the CIP circuit where activated carbon granules are mixed with the slurry and gold is adsorbed on to the activated carbon. The gold-loaded carbon is separated from the slurry and treated in an elution circuit to remove the gold. Comminution: Comminution is the crushing and grinding of ore to make gold available for physical or chemical separation (see also “Milling”). Contained gold or Contained copper: The total gold or copper content (tonnes multiplied by grade) of the material being described. Cut-off grade: Cut-off grade is the grade (i.e., the concentration of metal or mineral in rock) that determines the destination of the material during mining. For purposes of establishing “prospects of economic extraction,” the cut-off grade is the grade that distinguishes material deemed to have no economic value (it will not be mined in underground mining or if mined in surface mining, its destination will be the waste dump) from material deemed to have economic value (its ultimate destination during mining will be a processing facility). Other terms used in similar fashion as cut-off grade include net smelter return, pay limit, and break-even stripping ratio. Depletion: The decrease in the quantity of ore in a deposit or property resulting from extraction or production. Development: The process of accessing an orebody through shafts and/or tunneling in underground mining operations. Development stage property: A development stage property is a property that has Mineral Reserve disclosed, but no material extraction. Diorite: An igneous rock formed by the solidification of molten material (magma). Doré: Impure alloy of gold and silver produced at a mine to be refined to a higher purity. Economically viable: Economically viable, when used in the context of Mineral Reserve determination, means that the Qualified Person has determined, using a discounted cash flow analysis, or has otherwise analytically determined, that extraction of the Mineral Reserve is economically viable under reasonable investment and market assumptions. Electrowinning: A process of recovering gold from solution by means of electrolytic chemical reaction into a form that can be smelted easily into gold bars. Elution: Recovery of the gold from the activated carbon into solution before zinc precipitation or electrowinning. Exploration results: Exploration results are data and information generated by mineral exploration programs (i.e., programs consisting of sampling, drilling, trenching, analytical testing, assaying, and other similar activities undertaken to locate, investigate, define or delineate a mineral prospect or mineral deposit) that are not part of a disclosure of Mineral Resource or Reserve. A registrant must not use exploration results alone to derive estimates of tonnage, grade, and production rates, or in an assessment of economic viability. Exploration stage property: An exploration stage property is a property that has no Mineral Reserve disclosed. Exploration target: An exploration target is a statement or estimate of the exploration potential of a mineral deposit in a defined geological setting where the statement or estimate, quoted as a range of tonnage and a range of grade (or quality), relates to mineralisation for which there has been insufficient exploration to estimate a Mineral Resource.

AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 77 Feasibility Study (FS): A Feasibility Study is a comprehensive technical and economic study of the selected development option for a mineral project, which includes detailed assessments of all applicable modifying factors, as defined by this section, together with any other relevant operational factors, and detailed financial analysis that are necessary to demonstrate, at the time of reporting, that extraction is economically viable. The results of the study may serve as the basis for a final decision by a proponent or financial institution to proceed with, or finance, the development of the project. A Feasibility Study is more comprehensive, and with a higher degree of accuracy, than a Prefeasibility Study. It must contain mining, infrastructure, and process designs completed with sufficient rigor to serve as the basis for an investment decision or to support project financing. Flotation: Concentration of gold and gold-hosting minerals into a small mass by various techniques (e.g. collectors, frothers, agitation, air-flow) that collectively enhance the buoyancy of the target minerals, relative to unwanted gangue, for recovery into an over-flowing froth phase. Gold Produced: Refined gold in a saleable form derived from the mining process. Grade: The quantity of ore contained within a unit weight of mineralised material generally expressed in grams per metric tonne (g/t) or ounce per short ton for gold bearing material or Percentage copper (%Cu) for copper bearing material. Greenschist: A schistose metamorphic rock whose green colour is due to the presence of chlorite, epidote or actinolite. Indicated Mineral Resource: An Indicated Mineral Resource is that part of a Mineral Resource for which quantity and grade or quality are estimated on the basis of adequate geological evidence and sampling. The level of geological certainty associated with an Indicated Mineral Resource is sufficient to allow a qualified person to apply modifying factors in sufficient detail to support mine planning and evaluation of the economic viability of the deposit. Because an Indicated Mineral Resource has a lower level of confidence than the level of confidence of a Measured Mineral Resource, an Indicated Mineral Resource may only be converted to a Probable Mineral Reserve. Inferred Mineral Resource: An Inferred Mineral Resource is that part of a Mineral Resource for which quantity and grade or quality are estimated on the basis of limited geological evidence and sampling. The level of geological uncertainty associated with an Inferred Mineral Resource is too high to apply relevant technical and economic factors likely to influence the prospects of economic extraction in a manner useful for evaluation of economic viability. Because an Inferred Mineral Resource has the lowest level of geological confidence of all Mineral Resource, which prevents the application of the modifying factors in a manner useful for evaluation of economic viability. With caution AngloGold Ashanti uses Inferred Mineral Resource in its Mineral Reserve estimation process and the Inferred Mineral Resource is included in the pit shell or underground extraction shape determination. As such the Inferred Mineral Resource may influence the extraction shape. The quoted Mineral Reserve from these volumes includes only the converted Measured and Indicated Mineral Resource and no Inferred Mineral Resource is converted to Mineral Reserve. The cash flow analysis does not include the Inferred Mineral Resource in demonstrating the economic viability of the Mineral Reserve. Initial assessment (also known as concept study, scoping study and conceptual study): An initial assessment is a preliminary technical and economic study of the economic potential of all or parts of mineralisation to support the disclosure of Mineral Resource. The initial assessment must be prepared by a qualified person and must include appropriate assessments of reasonably assumed technical and economic factors, together with any other relevant operational factors, that are necessary to demonstrate at the time of reporting that there are reasonable prospects for economic extraction. An initial assessment is required for disclosure of Mineral Resource but cannot be used as the basis for disclosure of Mineral Reserve. Leaching: Dissolution of gold from crushed or milled material, including reclaimed slime, prior to adsorption on to activated carbon or direct zinc precipitation. Life of mine (LOM): Number of years for which an operation is planning to mine and treat ore, and is taken from the current mine plan. Measured Mineral Resource: A Measured Mineral Resource is that part of a Mineral Resource for which quantity and grade or quality are estimated on the basis of conclusive geological evidence and sampling. The level of geological certainty associated with a Measured Mineral Resource is sufficient to allow a qualified person to apply modifying factors, as defined in this section, in sufficient detail to support detailed mine planning and final evaluation of the economic viability of the deposit. Because a Measured Mineral Resource has a higher level of confidence than the level of confidence of either an Indicated Mineral Resource or an Inferred Mineral Resource, a Measured Mineral Resource may be converted to a Proven Mineral Reserve or to a Probable Mineral Reserve. Metallurgical plant: A processing plant constructed to treat ore and extract gold or copper in the case of Quebradona (and, in some cases, often valuable by-products). Metallurgical recovery factor (MetRF): A measure of the efficiency in extracting gold from the ore. Milling: A process of reducing broken ore to a size at which concentrating or leaching can be undertaken (see also “Comminution”). Mine call factor (MCF): The ratio, expressed as a percentage, of the total quantity of recovered and unrecovered mineral product after processing with the amount estimated in the ore based on sampling. The ratio of contained gold delivered to the metallurgical plant divided by the estimated contained gold of ore mined based on sampling. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 78 Mineral deposit: A mineral deposit is a concentration (or occurrence) of material of possible economic interest in or on the earth’s crust. Mining recovery factor (MRF): This factor reflects a mining efficiency factor relating the recovery of material during the mining process and is the variance between the tonnes called for in the mining design and what the plant receives. It is expressed in both a grade and tonnage number. Mineral Resource: A Mineral Resource is a concentration or occurrence of material of economic interest in or on the Earth's crust in such form, grade or quality, and quantity that there are reasonable prospects for economic extraction. A Mineral Resource is a reasonable estimate of mineralisation, taking into account relevant factors such as cut-off grade, likely mining dimensions, location or continuity, that, with the assumed and justifiable technical and economic conditions, is likely to, in whole or in part, become economically extractable. It is not merely an inventory of all mineralisation drilled or sampled. Modifying Factors: Modifying factors are the factors that a Qualified Person must apply to Indicated and Measured Mineral Resource and then evaluate in order to establish the economic viability of Mineral Reserve. A Qualified Person must apply and evaluate modifying factors to convert Measured and Indicated Mineral Resource to Proven and Probable Mineral Reserve. These factors include, but are not restricted to: Mining; processing; metallurgical; infrastructure; economic; marketing; legal; environmental compliance; plans, negotiations, or agreements with local individuals or groups; and governmental factors. The number, type and specific characteristics of the modifying factors applied will necessarily be a function of and depend upon the mineral, mine, property, or project. Ounce (oz) (troy): Used in imperial statistics. A kilogram is equal to 32.1507 ounces. A troy ounce is equal to 31.1035 grams. Pay limit: The grade of a unit of ore at which the revenue from the recovered mineral content of the ore is equal to the sum of total cash costs, closure costs, Mineral Reserve development and stay-in-business capital. This grade is expressed as an in-situ value in grams per tonne or ounces per short ton (before dilution and mineral losses). Precipitate: The solid product formed when a change in solution chemical conditions results in conversion of some pre-dissolved ions into solid state. Preliminary Feasibility Study (Prefeasibility Study or PFS): is a comprehensive study of a range of options for the technical and economic viability of a mineral project that has advanced to a stage where a qualified person has determined (in the case of underground mining) a preferred mining method, or (in the case of surface mining) a pit configuration, and in all cases has determined an effective method of mineral processing and an effective plan to sell the product. Probable Mineral Reserve: A Probable Mineral Reserve is the economically mineable part of an Indicated and, in some cases, a Measured Mineral Resource. Production stage property: A production stage property is a property with material extraction of Mineral Reserve. Productivity: An expression of labour productivity based on the ratio of ounces of gold produced per month to the total number of employees in mining operations. Project capital expenditure: Capital expenditure to either bring a new operation into production; to materially increase production capacity; or to materially extend the productive life of an asset. Proven Mineral Reserve: A Proven Mineral Reserve is the economically mineable part of a Measured Mineral Resource and can only result from conversion of a Measured Mineral Resource. Qualified Person: A Qualified Person is an individual who is (1) A mineral industry professional with at least five years of relevant experience in the type of mineralisation and type of deposit under consideration and in the specific type of activity that person is undertaking on behalf of the registrant; and (2) An eligible member or licensee in good standing of a recognised professional organisation at the time the technical report is prepared. Section 229.1300 of Regulation S-K 1300 details further recognised professional organisations and also relevant experience. Quartz: A hard mineral consisting of silica dioxide found widely in all rocks. Recovered grade: The recovered mineral content per unit of ore treated. Reef: A gold-bearing horizon, sometimes a conglomerate band, that may contain economic levels of gold. Reef can also be any significant or thick gold bearing quartz vein. Refining: The final purification process of a metal or mineral. Regulation S-K 1300: On 31 October 2018, the United States Securities and Exchange Commission adopted the amendment Subpart 1300 (17 CFR 229.1300) of Regulation S-K along with the amendments to related rules and guidance in order to modernise the property disclosure requirements for mining registrants under the Securities Act and the Securities Exchange Act. Registrants engaged in mining operations must comply with the final rule amendments (Regulation S-K 1300) for the first fiscal year beginning on or after 1 January 2021. Accordingly, the Company is providing disclosure in compliance with Regulation S-K 1300 for its fiscal year ending 31 December 2021 and will continue to do so going forward. AngloGold Ashanti La Colosa Project - 31 December 2021 _____________________________________________________________________________________ 30 March 2022 79 Rehabilitation: The process of reclaiming land disturbed by mining to allow an appropriate post-mining use. Rehabilitation standards are defined by country-specific laws, including but not limited to the South African Department of Mineral Resources, the US Bureau of Land Management, the US Forest Service, and the relevant Australian mining authorities, and address among other issues, ground and surface water, topsoil, final slope gradient, waste handling and re-vegetation issues. Resource modification factor (RMF): This factor is applied when there is an historic reconciliation discrepancy in the Mineral Resource model. For example, between the Mineral Resource model tonnage and the grade control model tonnage. It is expressed in both a grade and tonnage number. Scats: Within the metallurgical plants, scats is a term used to describe ejected ore or other uncrushable / grinding media arising from the milling process. This, typically oversize material (ore), is ejected from the mill and stockpiled or re-crushed via a scats retreatment circuit. Retreatment of scats is aimed at fracturing the material such that it can be returned to the mills and processed as with the other ores to recover the gold locked up within this oversize material. Seismic event: A sudden inelastic deformation within a given volume of rock that radiates detectable seismic energy. Shaft: A vertical or subvertical excavation used for accessing an underground mine; for transporting personnel, equipment and supplies; for hoisting ore and waste; for ventilation and utilities; and/or as an auxiliary exit. Smelting: A pyro-metallurgical operation in which gold precipitate from electro-winning or zinc precipitation is further separated from impurities. Stoping: The process of excavating ore underground. Stripping ratio: The ratio of waste tonnes to ore tonnes mined calculated as total tonnes mined less ore tonnes mined divided by ore tonnes mined. Tailings: Finely ground rock of low residual value from which valuable minerals have been extracted. Tonnage: Quantity of material measured in tonnes. Tonne: Used in metric statistics. Equal to 1,000 kilograms. Waste: Material that contains insufficient mineralisation for consideration for future treatment and, as such, is discarded. Yield: The amount of valuable mineral or metal recovered from each unit mass of ore expressed as ounces per short ton or grams per metric tonne. Zinc precipitation: Zinc precipitation is the chemical reaction using zinc dust that converts gold in solution to a solid form for smelting into unrefined gold bars. 25 Reliance on information provided by the Registrant The entire La Colosa study has been financed by the registrant. The relevant borehole information provided by the registrant (AngloGold Ashanti) has been extracted from the Century System Database and other data from the data room for the internal studies compiled by the QP in 2017. Reliance in information provided by the registrant includes guidance from the annual update to AngloGold Ashanti’s internal guidelines for reporting of Exploration Results, Mineral Resource and Ore Reserve. This guideline is set out to ensure the reporting of Exploration Results, Mineral Resource and Ore Reserve is consistently undertaken in a manner in accordance with AngloGold Ashanti’s business expectations and also in compliance with internationally accepted codes of practice adopted by AngloGold Ashanti. Included in this guideline is the price assumptions supplied by the Registrant which includes long-range commodity price and exchange rate forecasts. These are reviewed annually and are prepared in-house using a range of techniques including historic price averages. AngloGold Ashanti selects a conservative Mineral Reserve price relative to its peers. This is done to fit into the strategy to include a margin in the mine planning process. The resultant plan is then valued at a higher business planning price. The La Colosa Mineral Resource study was conducted in 2017 with a gold price of $1,400/oz.