July 11, 2008

Mr. Jim B. Rosenberg

Senior Assistant Chief Accountant

United States Securities and Exchange Commission

Division of Corporation Finance

100 F Street, N.E.

Washington, D.C. 20549

Re: Ambac Financial Group, Inc.

Form 10-K for the fiscal year ended December 31, 2007

Filed on February 29, 2008

Form 10-Q for the quarter ended March 31, 2008

Filed on May 12, 2008

File No. 1-10777

Dear Mr. Rosenberg:

This letter is submitted on behalf of Ambac Financial Group, Inc. (“Ambac”) in response to comments received from the Securities and Exchange Commission staff (“Staff”), in a letter dated June 26, 2008, with respect to Ambac’s Form 10-K for the fiscal year ended December 31, 2007 filed on February 29, 2008 (“2007 Form 10-K”) and Ambac’s Form 10-Q for the quarter ended March 31, 2008 filed on May 12, 2008 (“March 31, 2008 Form 10-Q”). The June 26, 2008 letter is a follow up to our responses to the SEC letter dated April 10, 2008, with respect to Ambac’s 2007 Form 10-K. The June 26, 2008 letter also included comments on Ambac’s March 31, 2008 Form 10-Q.

Ambac appreciates the efforts of the Staff in this review process. Enhancement of the overall disclosures in our filings is an objective that we share with the Staff and one that we continuously consider in our filings. In connection with responding to your comments, we acknowledge that Ambac is responsible for the adequacy and accuracy of the disclosures in our filings; that SEC staff comments or changes to disclosure in response to SEC staff comments do not preclude the SEC from taking any action with respect to the filing; and that Ambac may not assert SEC staff comments as a defense in any proceeding initiated by the SEC or any person under the federal securities laws of the United States.

For reference purposes, the text of your June 26, 2008 letter has been reproduced in this letter with responses directly following the reproduced text. The section subheader and page number in each comment refers to Ambac’s 2007 Form 10-K or March 31, 2008 Form 10-Q.

In each of our responses below where we indicate we will provide revised disclosures in the future, we will begin including those disclosures in the June 30, 2008 Form 10-Q.

Securities and Exchange Commission Staff Comments:

Form 10-K for the year ended December 31, 2007

Management’s Discussion and Analysis of Financial Condition and Results of Operations

Critical Accounting Policies and Estimates

Financial Guarantee Insurance Losses and Loss Expense, page 48

1. We acknowledge your response to prior comment 1 and note that the Second lien net loss reserve increased from $266 million at December 31, 2007 to $1.07 billion at March 31, 2008. Please revise your disclosure to quantify and discuss how management has adjusted each of the key assumptions (i.e. severity, prepayment rate estimation, and probability of default) used in

July 11, 2008

Page 2

calculating the current loss reserves for Second liens. This discussion should show the link between what has happened to the key assumptions in the past to what management is currently using as its key assumptions.

Ambac response:

We refer you to the last paragraph of page 34 of our March 31, 2008 Form 10-Q describing our loss estimate in the second lien credits segment of our non-derivative insurance portfolios. The pro-forma enhanced disclosure below, which has been black-lined to reflect changes, relates to our March 31, 2008 Form 10-Q. Comparable disclosure will be provided in future filings:

Second lien credits consist primarily of home equity line of credit (“HELOC”) and closed end second mortgage transactions. The probability of default and severity of loss for certain second lien mortgage transactions from the 2006 and 2007 vintages continued to exhibit deterioration in the first quarter of 2008, which resulted in a significant increase in reserves related to second lien credits from December 31, 2007 to March 31, 2008. In establishing our March 31, 2008 net loss reserves for second lien products we refined our approach to utilize a roll-rate methodology rather than the loss curve methodology employed at December 31, 2007. The following summary discusses in detail the assumptions used in the loss curve methodology for calculating December 31, 2007 loss estimates:

Loss Curve Methodology

• We modified a market default curve as the best representation we believed available of the rapid increase of front ended collateral losses evident in these transactions and the fairly rapid decrease of collateral losses that we assume to be the likely outcome;

• Related to the point above, we determined that a front ended loss curve with a rapid escalation culminating in peak defaults around month 21 should best capture the observed early default loss behavior in our underperforming transactions;

• After reaching peak level losses we assumed that default rates would decline for a subsequent period through about month 50 and then tail off for the balance of the transaction’s life;

• The database component of an industry standard analytic tool was used to retrieve actual collateral losses for each transaction; these actual losses were patterned to the shape of the loss curves discussed above in conjunction with deal specific constant prepayment rate (“CPR”) assumptions, which range between 8% and 18%, and observed market forward interest rates;

• The future pattern of collateral losses along the curve was used to determine input assumptions for the cash flow modeling component of the above-mentioned analytic tool, which then projected monthly losses;

• The resulting projections of monthly losses for each transaction were applied to the transactions’ individual capital structures in order to estimate claims on the Ambac insured tranches;

• The aggregated monthly claims estimates were then discounted to determine a present value loss estimate.

We refined our loss reserve approach for second lien transactions during the first quarter of 2008 for two primary reasons. First, the level of underperformance exhibited in our distressed exposures during the first quarter of 2008 has been unprecedented, making it difficult to forecast future losses using loss patterns based on market collateral performance similar to the collateral that was guaranteed, which is the basis of the loss curve methodology described above. Second, there have been significant differences in performance between pools with largely similar collateral characteristics.

Our loss estimates for the second lien products in the RMBS portfolio were determined using a roll rate methodology which was applied to each credit. The roll-rate methodology used to estimate the March 31, 2008 loss reserves for second-lien transactions, observes trends in delinquencies, defaults, loss severities, prepayments and extrapolates ultimate performance from this

July 11, 2008

Page 3

data. As more information (performance and other) accumulates we are able to update assumptions in this model to reflect these changes. By employing the roll-rate methodology, wWe examined the historical rate at which delinquent loans in each transaction rolled into later delinquency categories (i.e. 30-59 days, 60-89 days, 90+ days). We determined a pool specific current-to-30-to-59 day delinquency curve and applied a statistical regression technique to historical roll rates. We carried forward the non-performing mortgages through the delinquency pipeline through the 60-89 and 90+ delinquency categories all the way through to charge-off. Management reduced the transition rate between the 30-59 days and the 60-89 days categories by 50% after a plateau of 18 months, since we felt it was the best proxy for the current environment. Actual default rates escalated dramatically in the first quarter of 2008 with constant default rates (“CDRs”) on the subject transactions generally ranging between 15% and 25%. This data, along with the most recent delinquency information, was used to project a default curve for the life of the transaction. Projected prepayment rates utilized in the calculation of our reserve estimates were the greater of the average of the last several months’ prepayments, or six percent. We also used the lesser of 100% loss severities or historical loss severities if materially lower than 100%. We have not estimated the impact of representation and warranty breaches in individual transactions but are aggressively pursuing our remedies on each transaction. In future periods, and as information becomes available for us to estimate these loss mitigants, we will incorporate them into our overall loss estimates. These three metrics—the default curve, prepayment speed, and loss severity – are used in conjunction with an assumption that no clean-up call will be exercised to estimate monthly losses for each transaction. These monthly loss estimates were applied to the transactions’ individual capital structures in order to estimate claims on the Ambac insured tranches. The estimated claims were discounted at our current discount rate of 4.50% and reduced for any reinsurance we estimated to be recoverable to derive the net reserve of $1,066 million as of March 31, 2008. We have not estimated the impact of representation and warranty breaches in individual transactions but are aggressively pursuing our remedies on each transaction. In future periods, and as information becomes available for us to estimate these loss mitigants, we will incorporate them into our overall loss estimates.

Valuation of Financial Instruments, page 52

2. We acknowledge your response to prior comment 3. You state that the data vendor itself does not price the securities. Please disclose the techniques and the assumptions used to fair value these securities and the validation procedures you use to ensure the accuracy of the models used.

Ambac response:

As a point of clarification, the data vendor referred to does provide prices for the vast majority of securities in our investment portfolio. Such prices are accumulated from a variety of sources and compiled into a single data feed. That is, our vendor simply delivers the prices that are determined by other pricing sources. The prices provided from those sources represent professionally evaluated estimates of the amount that a buyer in the marketplace would pay for the security in a current sale. Since most fixed income securities do not trade on any given day, professional pricing sources generally must group securities by common factors. Appropriate market data such as yield curves and credit spreads are applied by the pricing sources in the valuation of securities within a given group and are updated daily based on applicable market and trade information.

As noted in our initial response to prior comment 3, the pricing sources used are well established in the fixed income market. Ambac performs various review and validation procedures to quoted prices, including: price variance analyses, missing and static price reviews, overall valuation analyses by senior traders and finance managers and reviews associated with our ongoing impairment analysis as disclosed in Note 2 to the Consolidated Financial Statements. Unusual prices identified through these procedures will be evaluated further against broker quotes (if available) or internally modeled prices, and the pricing source values will be challenged as necessary. Price challenges generally result in the use of the pricing source’s quote as originally provided or as revised by the

July 11, 2008

Page 4

source following their internal diligence process. A price challenge may result in a determination by the pricing source that they cannot provide a reasonable value for a security, in which case Ambac would resort to using either broker quotes or internal models. We will disclose these review and validation procedures in future filings.

3. We acknowledge your response to prior comment 5 and 7. It would appear that the distressed market environment and lack of CDS transactions is indicative of a change in market pricing. Please disclose why maintaining the same relative change ratio as existed at the CDS inception is appropriate when determining fair value at the measurement date in light of the current market conditions.

Ambac response:

As described in our response to prior comment 5, we have in fact adjusted the relative change ratio on transactions which have suffered credit downgrades. Below is an excerpt from our response to prior comment 5 (including changes black-lined against previous Form 10-K disclosures) which addresses question 3 above:

When reference obligations experience credit deterioration, there is an increase in the probability of default on the obligation and, therefore, an increase in expected loss. The effects of credit deterioration on financial guarantee CDS fees are not readily observable in the market. Ambac reflects the effects of changes in expected loss on the fair value of its CDS contracts by increasing the percentage of the reference obligation spread which would be captured as a CDS fee at the valuation date, resulting in a higher mark-to-market loss on our CDS relative to any price decline on the reference obligation. The fundamental assumption is that financial guarantee CDS fees will increase relative to reference obligation spreads as the underlying credit quality of the reference obligation deteriorates and approaches payment default. For example, if the credit spread of an underlying reference obligation was 80 basis points at the inception of a transaction and Ambac received a 20 basis point fee for issuing a CDS on that obligation, the “relative change ratio”, which represents the CDS fee to cash market spread Ambac would utilize in its valuation calculation, would be 25%. If the reference obligation spread increased to 100 basis points in the current reporting period, absent any observable changes in financial guarantee CDS market pricing or credit deterioration, Ambac’s current period CDS fee would be computed by multiplying the current reference obligation spread of 100 basis points by the relative change ratio of 25%, resulting in a 25 basis point fee. Thus, the model indicates we would need to receive an additional 5 basis points (25 bps currently less the 20bps we received at inception) for issuing a CDS in the current reporting period for this reference obligation. We would then discount the product of the notional amount of the CDS and the 5 basis point spread increase, over the weighted average life of the reference obligation to compute the current period mark-to-market loss. Using the same example, if the reference obligation spread increased to 100 basis points and there was credit deterioration as evidenced by an internal rating downgrade, we would adjust the relative change ratio upward and ultimately record a higher mark-to-market loss based on the computations described above absent any observable changes in financial guarantee CDS market pricing. We do not adjust the relative change ratio until an actual internal rating downgrade has occurred. However, because we have active surveillance procedures in place for our entire CDS portfolio, particularly for transactions at or near a below investment grade threshold, the likelihood that an internal downgrade would lag the actual credit deterioration of a transaction for any meaningful time period is considered unlikely.

In 2007, such adjustments to the relative change ratio were made on CDO of ABS transactions containing over 25% MBS exposure which have suffered credit downgrades. For Ambac’s 4 CDO of ABS transactions that are rated below investment grade (3 CDO squareds and one CDO of Mezzanine ABS), the average spread capture percentage ishas increased to approximately 89%. The factors used to increase the percentage of reference obligation spread captured in the CDS fee were based on recent rating agency probability of default percentages

July 11, 2008

Page 5

determined by management to be appropriate for structured finance CDO of ABS. Other asset types within the CDS portfolio have not experienced significant downgrades as of December 31, 2007 and as such do not reflect an adjustment to the relative change ratio. It is reasonably possible that additional downgrades could occur in the future on the CDO of ABS transactions in our CDS portfolio which contain over 25% MBS exposure. Assuming a one full letter downgrade on all of these transactions as of December 31, 2007, Ambac’s derivative liability balance would increase by $335 million.

In addition, when there are sufficient numbers of new observable transactions to indicate a general change in market pricing trends for CDS on a given asset class, management will adjust its assumptions about the percentage of reference obligation spreads captured as CDS fees to match the current market. No such adjustments were made in 2007. However, no CDS business is currently being transacted, some guarantors have stated they have exited this product, and it is possible insurance regulators will prohibit this product going forward.

4. We are continuing to evaluate your response and disclosures related to the fair value of your credit derivatives and we may have further comments.

Ambac response:

None.

RMBS Exposure in Collateralized Debt Obligations

Breakout of CDO of ABS>25% RMBS Exposure, page 59

5. We acknowledge your response to prior comment 10 and substantially reissue our comment. Please disclose the following as of the latest balance sheet date:

• The net derivative asset or liability of CDO of ABS>25% that is at a lower level than the total amount. Summarized by Ambac rating would be acceptable;

• The original amount of the subordination in your CDO of ABS when you entered into the credit default swap contracts, in this regard you may choose to disclose separately the amount of the subordination due to excess interest spread and the amount due to subordinated tranches separately;

• The amount of subordination in your CDO of ABS given that the amount of subordination is a key factor in determining the credit rating and

• The amount of distressed transactions in the process of remediation, the fair value of these transactions and the range of potential loss pending the restructuring of these transactions.

Ambac response:

With respect to the first bullet point above, below is the net derivative liability of CDO of ABS > 25% RMBS Exposure aggregated by Ambac rating as of December 31, 2007, which we will provide in future filings:

CDO of ABS > 25% RMBS Net Derivative Liability by Ambac Rating

at December 31, 2007 (in millions):

Ambac rating:	Net Derivative Liability:
AAA	$	(118	)
AA		(1,275	)
A		(1,189	)
BBB		(205	)
BIG		(2,148	)
Total	$	(4,935	)

July 11, 2008

Page 6

With respect to the second bullet point above, we acknowledge that original subordination may provide meaningful disclosure since it provides investors with the level of credit enhancement at the inception of a transaction. We have revised the table on page 59 of our 2007 Form 10-K to include the original amount of subordination in our CDO of ABS > 25% RMBS exposure, as shown in Appendix A, which we will provide in future filings. Please note that we have removed the Ambac credit ratings from this table which we will now include on the new disclosure table in Appendix B, further described below in our response to the third bullet point above.

With respect to the third bullet point above, we do not believe disclosing current subordination levels provides meaningful comparisons to original subordination levels for users of our financial statements because subordination of CDO of ABS transaction might exist only in nominal forms particularly when the credit quality of the underlying collateral has deteriorated significantly. In other words, subordinate bonds in a transaction structure continue to have a legal claim to the underlying collateral for any unpaid principal and interest, until such collateral pays down or is liquidated. Therefore the subordinate bonds are not technically “written down” to fair value nor are they reported as such in bond trustee reports. Additionally, if certain triggering events occur as a result of credit deterioration of the underlying collateral, cash flows from the underlying collateral often are diverted from the subordinate bonds to Ambac-insured senior bonds. In this situation, the outstanding par amount of the subordinate bonds as a percentage of the total par amount of senior and subordinate bonds, i.e. – the subordination level, has actually increased from the subordination level which existed at the origination of the transaction. Accordingly, in cases where the collateral in a structure has deteriorated significantly, reporting comparative subordination levels for the current period and at origination would be very misleading.

We believe providing the current credit quality of the underlying collateral, which ultimately supports the payment of obligations which Ambac has insured, would provide more meaningful disclosure than current subordination levels. As such, we propose providing the collateral table by rating shown in Appendix B, which relates to our December 31, 2007 Form 10-K exposures, for future filings.

With respect to the fourth bullet point above, all CDO of ABS transactions which are below investment grade (“BIG”) are currently in some stage of loss remediation. The fair values of those BIG transactions are shown in our pro-forma revised disclosure discussed in our response to the first bullet point above. We believe a reasonable range of potential losses for all CDO of ABS>25% RMBS transactions is between Ambac’s own estimate of credit impairment (established only for credits that are rated BIG) and S&P’s estimate of stress case losses. At December 31, 2007 Ambac’s estimate of credit impairment for its CDO of ABS exposures was $1.1 billion, which represents management’s estimate of expected future payments under our CDS. Based on information published in February 2008, S&P’s current estimate of stress case losses for Ambac’s CDO of ABS exposure was $3.7 billion on a present value basis. We will provide this added disclosure in future filings, which will include updated S&P stress numbers.

Liquidity and Capital Resources

Contractual Obligations, page 74

6. We have reviewed your response to prior comment 12 and note that you have not included credit reserves or credit impairment losses on credit default swaps in the contractual obligation table. It would appear that these liabilities represent future legal obligations of the Company. Due to the significant nature of these liabilities to your business we believe the inclusion of such amounts in the contractual obligation table will provide investors increased disclosure of your liquidity. The purpose of Financial Reporting Release 67 is to obtain enhanced disclosure concerning a registrant’s contractual payment obligations and the exclusion of ordinary course items would be inconsistent with the objective of the Item 303(a)(5) of Regulation S-K. Based on the above factors, please revise your contractual obligation table to include the expected settlement of your credit reserves and credit default swaps.

July 11, 2008

Page 7

Ambac response:

We will revise our contractual obligation table to also include the expected settlement of our active credit reserves and Ambac’s estimate of credit impairment on credit default swaps in future filings. We will also provide disclosure that these obligations are not scheduled obligations, are estimates and may change materially upon downgrades.

Risk Management

Market Risks, page 90

7. We acknowledge your response to prior comment 14 and 15. Please disclose the impaired credits and credit rating of the underlying assets for your “Other” and collateralized loan obligation derivative contracts as provided in your responses.

Ambac response:

As noted in our response to prior comments 14 and 15, we currently do not have any impaired credits in our “Other” and collateralized loan obligation derivative contracts. We will include those disclosures in future filings to the extent appropriate. We will also provide the credit rating of the underlying assets for those contracts in future filings.

Financial Statements

Note 10 – Income Taxes, page 128

8. We have reviewed your response to prior comment 18. You stated that the federal income tax treatment of credit default swap is an unsettled area of the tax law. You also stated you would be required to substantially reduce your deferred tax asset if the IRS decided that “pay-as-you-go” credit default swap contracts are capital assets. Please disclose the range of the substantial reduction in your deferred tax asset if the IRS characterized these contracts as capital assets. In addition, given the materiality and based on the uncertainty of your tax asset position as it relates to credit default swaps please disclose the technical merits and the factors that you considered in concluding that it was more-likely-than-not that credit default swap contracts could be treated as principal contracts as opposed to capital assets by the IRS.

Ambac response:

We will add the following disclosures to future filings:

The IRS has acknowledged that there is uncertainty with regards to the tax characterization of a credit default swap, and has indicated that it expects to issue guidance at some future date. Absent specific guidance, Ambac reached its own determination as follows:

Credit default swaps (“CDS contracts”) are financial instruments for which payments are determined by reference to some adverse “credit event” with respect to another financial “reference asset”. For federal tax purposes, CDS contracts are generally characterized as one of the following:

• Insurance – The credit protected party (“buyer”) enters into a policy with the credit risk protector (“seller”) under which the buyer agrees to pay fees in a lump sum or periodically to the seller. In return, the seller agrees to pay to the buyer the amount of any defaulted schedule payments on the reference asset. The contract provides that buyer must own the referenced asset and seller is subrogated to the buyer’s rights upon default.

July 11, 2008

Page 8

• Notional Principal Contract - The buyer enters into a credit default swap agreement under an ISDA standard form swap agreement with the seller, the terms of which coincide with the term of the reference asset. Buyer agrees to make periodic payments to the seller at specified intervals calculated by reference to objective financial information upon a notional principal amount for specified consideration or a promise to pay similar amounts.

• Put Option- The parties enter into a “ put option agreement” in which buyer pays a “put premium” in a lump sum or periodically to the seller in exchange for the right, exercisable upon a credit event, to sell to the seller the reference asset for its face amount. In such an event, the seller, as the new owner of the reference asset, has the right to proceed directly against the issuer of the reference asset. The credit event effectively terminates the contract.

Ambac began writing CDS contracts in 1999. Initially these contracts provided that upon a credit event, Ambac would pay buyer the face amount of the referenced asset in return for actual physical delivery of the referenced asset. These contracts were properly characterized as put options, with income on the premiums received deferred and recognized as capital gains when the contract lapses. As a capital asset, any losses on the contract were recognized as capital losses. This method of accounting for credit default swaps has been accepted by the IRS for calendar years 1999-2004.

Beginning in 2005, settlement of Ambac CDS contracts was altered to adopt a “pay as you go” settlement. The contracts call for the seller to make cash settlements to buyer in the event of failure of the referenced obligation to make any of a number of scheduled future payments. Cash settlement follows the payment schedule of the reference obligation in terms of amount and timing and does not terminate the contract. The conditions of settlement may be satisfied on any number of occasions. Buyer does not at anytime have any obligation to hold any part of the reference obligation, and is not required to transfer any obligation to seller at settlement. As such, the facts would indicate that “pay as you go” contracts should be characterized as notional principal contracts, with income recognized as premiums are received and losses recognized when paid. Additionally, both income and losses on notional principal contracts should be characterized as ordinary.

However, when Ambac began to write CDS contracts in a notional principal format (“pay as you go”) in 2005, the contracts continued to be treated as put options including deferring income/loss recognition until the contract lapsed and characterizing income/loss as a capital gain/loss. With regard to the timing of income recognition (deferred vs. upon receipt/payment), Ambac has filed for a “change of accounting method” with the IRS for the tax year 2008 in order to properly recognize premium income and losses when received and paid, respectively, rather than deferring. However, under current tax law Ambac must continue to defer income recognition from these CDS contracts on its tax filings until the IRS approves its “accounting method” change. With regard to the characterization of income/losses (capital vs. ordinary), the characterization of an item is a matter of fact which does not require IRS approval to change. Accordingly, since Ambac believes the facts clearly indicate that “pay as you go” contracts are notional principal contracts, as described above, we have properly re-characterized the income and losses on “pay as you go” contracts as ordinary. Nonetheless, in the event that the IRS were to ultimately decide that “pay as you go” contracts should be characterized as capital assets, a valuation allowance of approximately $2 billion would be required. The material losses in Ambac’s CDS portfolio relate to these “pay as you go” contracts.

9. In regards to prior comment 18, please disclose the following:

• The weight given to the various positive and negative indicators cited and the factors considered that support your conclusion that no valuation allowance is required.

• Clarify what you mean by an appropriate forecast of expenses in future periods and what actions you are required to make to achieve this forecast;

July 11, 2008

Page 9

• A more robust disclosure of the tax planning strategy that you have adopted and the projected impact of your tax strategy on ordinary income and capital gains given the negative evidence; and

• The impact that your recent downgrades by S&P and Moody’s will have on your valuation allowance analysis. Please tell us if the contracts for which future installment premiums are to be received are subject to possible cancellation by the holder as a result of the downgrades.

Ambac Response:

The following pro-forma disclosure below relates to our 2007 Form 10-K, and will be included in future filings. The disclosures below will be added to the disclosures we already discussed in our response to previous comment 18:

Ambac gave heavy weighting to the various negative indicators which raised concerns about its ability to write significant amounts of new business. Accordingly, in forecasting future taxable income, Ambac assumed that no new or renewal policies would be written. Significant weighting was given to the positive indicators of recognizing potential future premium earned on existing contracts (i) in which the premium was paid up front and (ii) in which the premium is anticipated in the future . The amount of installment premiums actually realized could be reduced in the future due to factors such as early termination of the contracts or accelerated prepayments of underlying obligations. Ambac regularly monitors terminations and accelerated payments and will reflect any impact in future quarters. The recent downgrade by S&P and Moody’s is not expected to have a material impact on future installment premium collections. Heavy weighting was also given to Ambac’s ability to generate future investment income on existing assets as adjusted for anticipated future cash flows, assuming a return equal to its current return on assets. Ambac further recognized that future taxable income could be increased by selling its municipal portfolio and reinvesting in taxable securities. Although its municipal portfolio was in an unrealized gain position and taxable securities were yielding higher than municipal securities, presently, Ambac did not recognize any capital gain income or increased asset yield as a result of this strategy.

With respect to underwriting and operating expense, it was assumed that under a scenario in which no new business would be written, future operating expenses would be significantly reduced. Ambac assumed a 50% reduction in expenses beginning in 2010 related predominantly to reduced compensation costs, with a 5% annual growth rate for inflation.

The scheduling of the various projected future income and expense indicated an excess of $5 billion in anticipated future income over anticipated future losses leading Ambac to conclude that no valuation allowance was required.

Note 19 – Segment Reporting, page 146

10. We have reviewed your response to prior comment 19. It is unclear why you have only two operating segments. Please address the following:

• Clarify what you mean by multiple levels of discrete financial information, and describe this discrete financial data in sufficient detail and how this information is broken out by senior business managers;

• According to a subsidiary organizational chart published in a recent Fitch Ratings report Ambac Credit Products, Ambac Assurance U.K. Ltd, Connie Lee Holdings and Ambac Financial Services are wholly owned subsidiaries of Ambac Assurance Corp., and Ambac Capital Funding and Ambac Investments are wholly owned subsidiaries of Ambac Capital Corp. In your response you state that Ambac Financial Services represents one segment and imply that the other five subsidiaries are aggregated in the financial guarantee segment. Please tell us why the various subsidiaries identified above do not meet the definition of an operating segment based on the guidance in paragraph 10 of SFAS 131. Please also reconcile the definition of an operating segment as defined in paragraph 10 of SFAS 131 and your response that you only have two operating segments;

July 11, 2008

Page 10

• Tell us the financial measures or ratios used by the chief operating decision maker to evaluate financial performance of the segments;

• On page 36 of the proxy statement you filed recently you disclosed that compensation in Executive Incentive Plan is based on market position/new business production and new products/initiatives and risk management. Since the performance formula for long-term compensation is based on these factors among others it would appear that you would need to generate discrete data for your high growth and high risk credit derivative products. Please tell us what the performance formula for long-term compensation has been for the last three fiscal years and the current fiscal year. Please also describe the financial and management reports used by senior management and the Compensation Committee to evaluate performance and achievement of market position/new business production, new products/initiatives and risk management goals; and

• Provide us with a summary of the kind of data that was provided to the chief operating decision maker for the 2007 fiscal year and through the current interim period.

Ambac response:

With respect to the first bullet point, by “multiple levels of discrete financial information” we mean the two ways in which discrete financial information is provided to the Chief Executive Officer on a monthly basis. First, discrete financial information is segregated by types of products, which are financial guarantee products (financial guarantee and credit default swaps in the aggregate) and financial services products (investment agreements, funding conduits and derivatives in the aggregate). Second, for the financial guarantee business, discrete financial information is further broken out by five business groups within an internally prepared monthly financial report (“MFR”) as shown below:

Group 1: West Region Public Finance and North America Project Finance

Group 2: East Region Public Finance; Healthcare; Structured Real Estate

Group 3: Australia/Pacific; Europe; Japan; Milan

Group 4: Asset-backed Securities, Conduits, Emerging Markets, and Utilities

Group 5: Structured Credit

Note that each senior business manager might be responsible for more than one business group and there are overlaps in asset classes, geographic regions and execution form (i.e. financial guarantee policy vs. credit default swap) amongst the business groups. For example, the business manager responsible for Utility risks would include both financial guarantee policy and credit default swap execution.

The MFR broken out by each business group includes the following key performance indicators: insured notional, average premium rate, premium written, net premium earned (net of reinsurance). Additionally, financial statement amounts are allocated to each business manager such as investment income, loss expenses, underwriting expenses, taxes, interest expense and net earnings.

With respect to the second bullet point we want to clarify in our response to prior comment 19 that when we refer to “Financial Services”, this refers to a reportable segment that is not the same as Ambac Financial Services LLC, which is a subsidiary of Ambac Assurance Corp. (“AAC”).

Financial Services segment - The Financial Services segment includes business activities from the following subsidiaries: Ambac Financial Services LLC (derivatives), Ambac Capital Services LLC (derivatives), and Ambac Capital Corporation and subsidiaries (investment agreements and funding conduits). Ambac Capital Funding and Ambac Investments are wholly-owned subsidiaries of Ambac Capital Corporation. Ambac Financial Services LLC and Ambac Capital Services LLC are wholly- owned subsidiaries of AAC. Collectively, these subsidiaries provide financial services products consisting of derivatives, investment agreements and funding conduits that are complementary in

July 11, 2008

Page 11

nature to our core financial guarantee insurance product discussed in the next section. However, all of these legal entities fail criteria (b) in paragraph 10 of SFAS 131 to qualify as individual operating segments because the Chief Executive Officer does not regularly review the operating results at the legal entity level. Rather, discrete financial information for the Financial Services segment as a whole is reviewed by the Chief Executive Officer on a regular basis for the purposes of making decisions about resources to be allocated or to assess their performances.

Financial Guarantee segment - Ambac Credit Products (credit default swaps), Ambac Assurance U.K. Ltd (financial guarantee) and Connie Lee Holdings (financial guarantee policy), are wholly-owned subsidiaries of AAC. Collectively these subsidiaries, including AAC, provide our financial guarantee product. The credit default swap product provided by Ambac Credit products is substantively the same as the financial guarantee product provided by the other subsidiaries and differs only in form of transaction execution. However, all of these legal entities fail criteria (b) in paragraph 10 of SFAS 131 to qualify as individual operating segments because the Chief Executive Officer does not regularly review the operating results at the legal entity level. As discussed in our response to the first bullet point above, the operating results of each business group are aggregated for review by the Chief Executive Officer, and there are overlaps in asset classes, geographic regions and execution form (i.e. financial guarantee policy vs. credit default swap) amongst the business groups.

Consequently, management has concluded that there are two reportable segments: Financial Services and Financial Guarantee.

With respect to the third bullet point, refer to our response to the first bullet point for the summary of financial measures provided to the Chief Executive Officer at the Financial Guarantee segment level. In addition, the Chief Executive Officer is provided the following at the reportable segment level:

• GAAP Revenues

• GAAP Expenses

• Pre-tax GAAP income

• Return on internally allocated capital (Financial Guarantee segment only)

With respect to the fourth bullet point, Ambac’s Executive Incentive Plan (“EIP”), as outlined on page 36 of the 2008 Proxy Statement, provides a performance formula for calculating the maximum incentive compensation payable under the EIP. The incentive compensation can be paid in cash and/or equity awards and Ambac’s incentive compensation has been comprised of cash bonus, restricted stock units and stock options (the latter of which is not subject to the EIP).

No later than 90 days after the beginning of the performance year, the Compensation Committee establishes a performance based formula for calculating the maximum incentive compensation payable under the EIP. The performance formula in the EIP may be based on any one or any combination of the performance categories set forth in the Proxy and, for 2007, the Committee selected return on equity.

There is no discreet data for “high growth and high risk credit derivative products” that is generated for purposes of determining long-term compensation under the EIP. The Committee is presented with information with regard to the performance formula and other information which is catalogued immediately below:

• Compensation Component Comparison for each group by compensation type and in the aggregate

• Year on year comparison of totals for all forms of compensation (company-wide)

• Year on year bonus and other company-wide performance metrics

• Proxy Comparison data, Peer Performance Comparison data and Market Pricing for Named Executive Officers

• Executive Officer Total Compensation analysis and proposal

July 11, 2008

Page 12

• Stock Ownership Impact on Executive Officers

• Executive Incentive Plan

• Chairman’s Report and Corporate Performance Scorecard (2005 and 2006)

• Summary if CEO’s Total Compensation and Review of Historical Awards, CEO Annual Compensation Tally Sheet, (2005 and 2006)

The Committee considers total compensation and the appropriate balance of cash and long term incentive compensation in an attempt to maintain alignment between the Executive and our shareholders. The CEO also answers any questions of the Committee members in response to his recommendations for long term (or other forms of) compensation.

The performance formula under the EIP for the last three fiscal years is as follows:

2007 Performance Year—Return on Equity

2006 Performance Year—Return on Equity

2005 Performance Year—Return on Equity x Core Earnings x 2 = Bonus %

Bonus % x 2005 Base Salary = Bonus

The performance formula under the EIP for the current fiscal year is driven by core earnings.

With respect to the fifth bullet point, refer to our response to the third bullet point above.

Form 10-Q for the quarter ended March 31, 2008

Note 9 – Fair Value Measurements, page 20

11. You classified all of your fixed-income securities as either Level 1 or Level 2 in the SFAS 157 hierarchy. However, you disclosed in your recent annual report that trading volume in residential mortgage-backed and certain asset-backed securities has been extremely limited. Please tell us the factors that you considered in classifying illiquid and structurally customized mortgage and asset-backed securities as Level 2 securities as opposed to Level 3 securities.

Ambac response:

Within our disclosures of critical accounting policies, we explain that nearly all of the fair values of our fixed income investments reflect quotes received either from a dealer or alternative pricing sources. Among the investments valued using such quotes are residential mortgage and certain asset-backed securities. We believe it is prudent to highlight to users of our financial statements that trading volume in the markets for such securities has been extremely limited, which could reasonably result in substantial variances in quoted prices over time or between different pricing sources. That said, we have continued to obtain quotes for these securities from our traditional sources. We believe these quotes are consistent with the inputs characterized as Level 2 under FAS 157. Specifically, paragraph 28.b. of FAS 157 states that Level 2 inputs include “quoted prices for identical or similar assets or liabilities in markets that are not active, that is, markets in which there are few transactions for the asset or liability, the prices are not current, or price quotations vary substantially either over time or among market makers.” Conversely, we do not believe these assets should be considered Level 3 as the valuation process did not rely on any of our own assumptions or data.

July 11, 2008

Page 13

Residential Mortgage-Backed Securities Exposure

Direct RMBS Portfolio Exposure, page 40

12. The credit quality of your mid-prime residential mortgage-backed security appears to have deteriorated significantly during the three months ended March 31, 2008. For example, AAA rated mid-prime securities declined from over 78% of your mid-prime portfolio at December 31, 2007 to 46% at March 31, 2008; your AA rated mid-prime securities declined from over 18% of your mid-prime portfolio at year-end to 0% at March 31, 2008; below investment grade securities increased from 0% of your mid-prime portfolio at year-end to almost 24% at March 31, 2008 and you had a substantial increase in BBB rated securities during the quarter. Please provide a robust disclosure of the factors that contributed to the significant decrease in credit quality in your mid-prime portfolio in the 90 days since year-end. As part of this discussion please be sure to clarify why mid-prime securities rated AA or higher decreased from almost 97% of your portfolio at year-end to 46% of your portfolio at March 31, 2008.

Ambac response:

The Alt-A portion of our RMBS portfolio is classified as “mid-prime” for purposes of the tables shown on page 41 of the March 31, 2008 Form 10-Q. We refer you to pages 34 and 35 of the March 31, 2008 Form 10-Q, which we believe provides robust disclosure regarding the factors which contributed to the significant credit deterioration in the Alt-A portfolio during the first quarter of 2008. We will clarify the definition of “mid-prime” exposure and clearly state that it includes Alt-A transactions for future filings.

13. Given that 24% of your mid-prime mortgage portfolio rated below investment grade was underwritten between 2005 through 2007 please disclose why you have not recorded any estimated probable losses on your mid-prime residential mortgage-backed security guarantees.

Ambac response:

As noted in our response to question 12 above, the mid-prime portfolio includes Alt-A transactions, which we will clarify in future filings. We refer you to the table on page 34 of our March 31, 2008 Form 10-Q, which indicates that we have recorded $200 million in net loss reserves for Alt-A RMBS transactions. Thus, we have in fact recorded estimated probable losses on our mid-prime portfolio during the first quarter of 2008.

Item 3. Quantitative and Qualitative Disclosure About Market Risk

Market Risk, page 77

14. We note that the input of your credit default swap spread into your fair value calculation resulted in a $1.6 billion reduction in your credit derivative liability. Please provide a sensitivity analysis for your credit default swap spreads similar to the one provided on page 80 for market credit spreads.

July 11, 2008

Page 14

Ambac response:

The pro-forma disclosure below relates to our March 31, 2008 Form 10-Q, which we will include in future filings:

The following table summarizes the estimated change in fair values on the net balance of Ambac’s structured credit derivative and total return swap positions assuming immediate shifts in Ambac Assurance’s credit spread at March 31, 2008 ($ in millions):

	Estimated Unrealized Gain / (Loss)
Change in Ambac Assurance credit spreads	CDO of ABS>25% RMBS											Total Estimated Unrealized Gain(Loss)
	CDO of ABS			CDO of CDO		CLO		Other		Total
2000 basis point widening	$	1,752		648		183		111		2,694		$	(5,090	)
1000 basis point widening		1,036		383		110		66		1,595			(6,189	)
500 basis point widening		569		210		61		37		877			(6,907	)
Base scenario		—		—		—		—		—			(7,784	)
500 basis point narrowing		(702	)	(258	)	(78	)	(46	)	(1,084	)		(8,868	)
1000 basis point narrowing		(1,053	)	(385	)	(110	)	(68	)	(1,616	)		(9,400	)
2000 basis point narrowing		(1,053	)	(385	)	(110	)	(68	)	(1,616	)		(9,400	)

Please feel free to contact me at (212) 208-3177 should you require further information or have any questions.

Sincerely,

Sean T. Leonard

Senior Vice President and

Chief Financial Officer

Copy to:	Michael Callen
	Kevin J. Doyle, Esq.
	John Schwolsky, Esq
	Dewey LeBoeouf

July 11, 2008

Page 15

Joel Parker

Accounting Branch Chief

Securities and Exchange Commission

Gus Rodriguez

Staff Accountant

Securities and Exchange Commission

John Bendl

KPMG

Paul Laurenzano

KPMG

Appendix A

Breakout of CDO of ABS >25% RMBS Exposure⁽¹⁾

			Collateral as % of Deal(2)
Year Issued	CDO of ABS	Amount(1) (as of 12/31/07) ($ millions)	Sub-prime RMBS(3)		Other RMBS(4)		ABS CDO High-grade		ABS CDO Mezzanine		CDO Other(5)		Other ABS(5)		Original Subordination(6)
2004	Cheyne High Grade ABS CDO, Ltd.	780	37	%	16	%	10	%	11	%	24	%	2	%	22	%
2005	Duke Funding High Grade III Ltd.	1,524	37	%	62	%	—		—		—		—		17	%
2005	Palmer Square PLC	988	36	%	33	%	7	%	7	%	15	%	1	%	21	%
2005	Hereford Street ABS CDO I, Ltd.	986	56	%	17	%	—		—		23	%	4	%	17	%
2005	Pascal CDO, Ltd.	859	58	%	17	%	3	%	5	%	6	%	10	%	14	%
2005	Tremonia CDO 2005-1 PLC	815	43	%	27	%	2	%	10	%	13	%	5	%	18	%
2005	High Grade Structured Credit CDO 2005-1	624	59	%	31	%	3	%	5	%	2	%	—		17	%
2005	Belle Haven ABS CDO 2005-1, Ltd.	588	56	%	30	%	4	%	4	%	2	%	3	%	22	%
2006	Diversey Harbor ABS CDO, Ltd.	1,868	35	%	40	%	7	%	17	%	1	%	—		23	%
2006	Belle Haven ABS CDO 2006-1, Ltd.	1,667	45	%	29	%	5	%	9	%	5	%	7	%	15	%
2006	Ridgeway Court Funding I, Ltd.	1,567	36	%	26	%	8	%	23	%	4	%	2	%	20	%
2006	Duke Funding High Grade IV, Ltd.	1,303	31	%	68	%	—		—		—		—		13	%
2006	Duke Funding High Grade V, Ltd.	1,250	39	%	61	%	—		—		—		—		16	%
2006	McKinley Funding III, Ltd.	1,179	18	%	40	%	6	%	32	%	3	%	—		19	%
2006	Millerton II High Grade ABS CDO, Ltd.	1,116	41	%	51	%	3	%	1	%	3	%	—		14	%
2006	Lancer Funding, Ltd.	940	47	%	41	%	3	%	6	%	3	%	1	%	20	%
2006	Cairn High Grade ABS CDO II Limited	819	34	%	41	%	1	%	21	%	1	%	2	%	18	%
2006	ESP Funding I, Ltd.	670	44	%	15	%	—		17	%	22	%	—		28	%
2006	Longshore CDO Funding 2006-1, Ltd.	614	30	%	32	%	7	%	12	%	16	%	2	%	17	%
2007	Kleros Preferred Funding VI, Ltd.	2,397	34	%	35	%	6	%	20	%	3	%	1	%	20	%
2007	Ridgeway Court Funding II, Ltd.	1,942	48	%	10	%	3	%	31	%	7	%	1	%	35	%
2007	Citation High Grade ABS CDO I, Ltd.	937	39	%	53	%	—		5	%	—		2	%	15	%
2007	Fiorente Funding Limited	721	35	%	42	%	2	%	12	%	1	%	7	%	15	%
2007	CDO of Mezzanine ABS	503	83	%	6	%	—		7	%	—		4	%	49	%
	Subtotal	26,655
	CDO of CDO
2005	CDO of Mezzanine ABS	79	9	%	1	%	20	%	33	%	36	%	1	%	50	%
2007	CDO of Mezzanine ABS	1,399	—		—		—		100	%	—		—		30	%
2007	CDO of Mezzanine ABS	497	—		—		7	%	93	%	—		—		48	%
2007	CDO of Mezzanine ABS	498	—		—		—		100	%	—		—		48	%
	Subtotal	2,472
	Total	29,127

(1) Amounts exclude an outstanding commitment with respect to approximately $2.9 billion of ABS CDOs which carries an Ambac rating of BBB+. This commitment is disclosed in further detail below in “Other CDO Commitments”.

(2) May not total 100% due to rounding.

(3) “Subprime”—Generally, transactions were categorized as sub-prime if they had a weighted-average credit score of 640 or lower as set forth in a third party data source Ambac deemed reliable. If no credit score was available, transactions were categorized as sub-prime unless a reliable third party source, such as a rating agency, categorized the transaction as prime or mid-prime.

(4) “Other RMBS”—Generally, transactions were categorized as Other RMBS if they were RMBS transactions with a weighted-average credit score greater than 640 (at origination of mortgages) as set forth in a third party data source we deemed reliable. If no credit score was available, transactions were categorized as Other RMBS if a reliable third party source, such as a rating agency, categorized the transaction as prime or mid-prime.

(5) “CDO Other” and “Other ABS”—Generally, transactions were categorized as either “CDO Other” or “Other ABS” if they are not High-grade CDO of ABS or Mezzanine CDO of ABS, Subprime or Other RMBS as described above. Examples of types of transactions included in the CDO Other category include, but are not limited to; CDOs primarily backed by commercial MBS or corporate securities, and collateralized loan obligations. Examples of transactions included in the Other ABS category, include, but are not limited to, transactions backed by commercial MBS, student loans, automobile loans, credit card receivables and student loans.

(6) Original subordination means, with respect to each CDO of ABS, the total subordination below Ambac as of the related issuance date.

Appendix B

Ratings of Underlying Collateral of CDO of ABS >25% RMBS Exposure

Year Issued	CDO of ABS	Amount (as of 12/31/07) ($ millions)	AAA(1)		AA(1)		A(1)		BBB(1)		BIG(1)		Ambac Rating(2)
2004	Cheyne High Grade ABS CDO, Ltd.	780	37	%	48	%	12	%	3	%	—		AA+
2005	Duke Funding High Grade III Ltd.	1,524	2	%	53	%	36	%	6	%	3	%	A-
2005	Palmer Square PLC	988	37	%	56	%	7	%	—		—		AAA
2005	Hereford Street ABS CDO I, Ltd.	986	10	%	29	%	35	%	24	%	2	%	AA+
2005	Pascal CDO, Ltd.	859	19	%	57	%	22	%	2	%	—		AAA
2005	Tremonia CDO 2005-1 PLC	815	20	%	35	%	41	%	1	%	3	%	AAA
2005	High Grade Structured Credit CDO 2005-1	624	6	%	28	%	60	%	5	%	1	%	AA+
2005	Belle Haven ABS CDO 2005-1, Ltd.	588	6	%	2	%	34	%	45	%	13	%	AA+
2006	Diversey Harbor ABS CDO, Ltd.	1,868	25	%	34	%	24	%	6	%	11	%	AA-
2006	Belle Haven ABS CDO 2006-1, Ltd.	1,667	19	%	38	%	27	%	2	%	14	%	A
2006	Ridgeway Court Funding I, Ltd.	1,567	26	%	38	%	21	%	4	%	11	%	A
2006	Duke Funding High Grade IV, Ltd.	1,303	1	%	44	%	45	%	9	%	1	%	AA-
2006	Duke Funding High Grade V, Ltd.	1,250	—		33	%	45	%	15	%	7	%	A+
2006	McKinley Funding III, Ltd.	1,179	27	%	44	%	12	%	5	%	12	%	BBB-
2006	Millerton II High Grade ABS CDO, Ltd.	1,116	24	%	39	%	32	%	2	%	3	%	AA-
2006	Lancer Funding, Ltd.	940	14	%	50	%	30	%	3	%	3	%	AA-
2006	Cairn High Grade ABS CDO II Limited	819	8	%	44	%	35	%	10	%	3	%	A+
2006	ESP Funding I, Ltd.	670	19	%	45	%	27	%	5	%	4	%	A+
2006	Longshore CDO Funding 2006-1, Ltd.	614	27	%	32	%	29	%	4	%	8	%	AA+
2007	Kleros Preferred Funding VI, Ltd.	2,397	48	%	19	%	15	%	7	%	11	%	A
2007	Ridgeway Court Funding II, Ltd.	1,942	5	%	36	%	22	%	14	%	23	%	AA-
2007	Citation High Grade ABS CDO I, Ltd.	937	30	%	34	%	21	%	6	%	9	%	A+
2007	Fiorente Funding Limited	721	30	%	44	%	10	%	4	%	12	%	AA-
2007	CDO of Mezzanine ABS	503	—		1	%	1	%	10	%	88	%	BIG
	Subtotal	26,655

Year Issued	CDO of ABS	Amount (as of 12/31/07) ($ millions)	AAA(1)		AA(1)		A(1)		BBB(1)		BIG(1)		Ambac Rating(2)
	CDO of CDO
2005	CDO of Mezzanine ABS	79	—		—		1	%	80	%	19	%	BBB-
2007	CDO of Mezzanine ABS	1,399	—		28	%	26	%	14	%	32	%	BIG
2007	CDO of Mezzanine ABS	497	1	%	2	%	17	%	22	%	58	%	BIG
2007	CDO of Mezzanine ABS	498	—		—		10	%	24	%	66	%	BIG
	Subtotal	2,472
	Total	29,127

(1) The ratings set forth above are as of the date specified, and may be changed at any time by the rating agencies. Ambac undertakes no obligation to update such ratings.

(2) Internal Ambac credit ratings are provided solely to indicate the underlying credit quality of guaranteed obligations based on the view of Ambac. Ambac ratings set forth above reflect the internal Ambac ratings as of December 31, 2007, and may be changed at any time based on our internal credit review. Ambac undertakes no obligation to update such ratings. This does not constitute investment advice. Ambac or one of its affiliates has guaranteed the obligations listed and may also provide other products or services to the issuers of these obligations for which Ambac may have received premiums or fees. Ambac undertakes no obligation to update ratings. “BIG” denotes credits deemed below investment grade (e.g, below BBB-).