MEMORANDUM
JED S. RAKOFF, District Judge.Plaintiff Assured Guaranty Municipal Corporation (“Assured”)1 brings this ac*599tion alleging that defendants Flagstar Bank, FSB; Flagstar Capital Markets Corporation; and Flagstar ABS, LLC (collectively, “Flagstar”) breached a series of contracts between the parties relating to financial guaranty insurance on nearly $1 billion in Flagstar securities that were backed by home equity loans. Assured alleges that Flagstar made a number of false representations with respect to those loans, thus breaching express representations and warranties Flagstar made in the Transaction Documents for two transactions — the 2005-1 transaction and the 2006-2 transaction (collectively the “Transactions”) — and that Flagstar breached its duties as a servicer with respect to the Transactions. Following discovery, Flags-tar moved for summary judgment. Upon consideration, this Court, on February 29, 2012, issued a bottom-line order denying Flagstar’s motion for summary judgment in its entirety. This Memorandum explains the reasons for that ruling.
The pertinent facts, either undisputed or where genuinely disputed viewed in the light most favorable to the plaintiff, are as follows. Assured provides bond insurance for, among other things, residential mortgage backed securities (“RMBS”). As a bond insurer, Assured guarantees timely payment of interest and principal to bondholders. Statement of Undisputed Material Facts Pursuant to Local Rule 56.1 of Defendants (“Def. 56.1”) ¶¶ 1-2; Plaintiffs Local Rule 56.1 Statement of Material Facts (“PI. 56.1”) ¶¶ 1-2. In the Transactions, Assured agreed to provide Flagstar with financial guaranty insurance for approximately $1 billion worth of Flagstarissued securities, which were backed by several thousand home equity loans.2 Plaintiffs Statement of Additional Material Facts, (“PI. Reply 56.1”) ¶ 1; Reply Statement of Undisputed Material Facts Pursuant to Local Rule 56.1 of Defendants (“Def. Reply 56.1”). The parties’ agreements were memorialized in a set of three simultaneously executed contracts: the Sale and Servicing Agreements (the “SSAs”), the Mortgage Loan Purchasing Agreements (the “MLPAs”), and the Insurance and Indemnity Agreements (the “I & I’s”) (collectively, the “Transaction Documents”).3
Flagstar made numerous representations and warranties in the MLPAs regarding the quality and characteristics of the underlying mortgages, including that: (1) “Each Mortgage Loan was originated in good faith and in accordance with the [Flagstar’s] underwriting guidelines;” (2) “[n]o error, omission, misrepresentation, negligence, fraud or similar occurrence with respect to a Mortgage Loan has taken place on the part of any person;” and (3) at “the time of origination [Flagstar] had no knowledge of any fact that would have led it to expect that any interest in any Mortgage Loan is unlikely to be paid in full when it becomes due and payable.” See Declaration of Veronica E. Rendon, dated May 23, 2011 (“Rendon May Deck”) Ex. M. (“MLPA”) § 3.02(a).
The SSAs designate Assured as a third-party beneficiary, and they repeat and incorporate Flagstar’s representations and warranties from the MLPAs. See Rendon May Deck, Ex. N (“SSA”) §§ 2.04(a), 8.06. Furthermore, the SSAs provide that Flagstar is liable for any material breach *600of those warranties, regardless of whether or not Flagstar knew that “[their] substance ... was inaccurate at the time” they were made. See id. § 2.04(b).
The I & I’s, in turn, incorporate the representations and warranties made by Flagstar in the SSAs and MLPAs; they also contain several additional representations and covenants by Flagstar, including that: (1) “[N]one of the [Transaction] Documents contain any statement of a material fact with respect to ... the Mortgage Loans that was untrue or misleading in any material respect,” and (2) “[t]he information supplied by [Flagstar] to [independent ratings agencies] did not contain any untrue statement of a material fact or omit to state any material fact.” Declaration of Jacob Buchdahl, Jan. 23, 2012 (“Buchdahl Deck”) Ex. A (“I & I”) § 2.01. As an explicit “condition precedent” to the issuance of the policies, Flagstar certified that “[t]he representations and warranties of [Flagstar] in this Agreement ... shall be true and correct in all material respects.” Id. App’x. A at 1.
Flagstar’s 30(b)(6) witness Matthew Roslin, Flagstar’s chief legal officer and the signatory to all of the contracts, stated that the securitizations were comprised of “for lack of a better term, the kitchen sink of any HELOCs that we had.” PI. Reply 56.1 ¶ 25; Def. Reply 56.1 ¶25. Roslin also testified that it is typical for underwriting errors that trigger repurchase to occur in transactions like the ones at issue here. He said, “[f]or years and years and years, certainly as long as I’ve been in the industry, there are repurchases in the normal course. There are errors made. There are origination errors that warrant repurchase.” PI. Reply 56.1 ¶28; Def. Reply 56.1 ¶ 28. Roslin went on to observe that “[t]he nature of the [origination] process is not perfect and there are bound to be some loans that we would have anticipated in our experience are going to have an origination error of some type.” PI. Reply 56.1 ¶ 29; Def. Reply 56.1 ¶ 29.
Flagstar was aware that there were some origination problems in the 2005 and 2006 transaction loan pools, Buchdahl Deck, Ex. D, Deposition of Jean Garrick (“Garrick Dep.”) at 19:12-20:5; 56:1-57:18. Flagstar was also aware that there was some fraud in the loan pools. It referred over 100 loans for review by its fraud investigation unit, including seven in the 800-loan sample that Assured is using in this case. Of those seven, Assured’s rewriting found that each contained serious breaches, including four that involved fraud. Buchdahl Deck, Ex. H. Roslin testified, however, that there was no “loan-by-loan analysis done after the origination process.” PI. Reply 56.1 ¶ 26; Def. Reply 56.1 ¶ 26.
For purposes of this litigation, Assured created two random samples of 400 loans from each of the two transactions, for a total of 800 loans. Def. 56.1 ¶ 26; PI. 56.1 ¶ 26. Plaintiffs expert, Rebecca Walzak, reviewed those 800 loans, and she concluded that 610 of those loans contain breaches of Flagstar’s representations and warranties. Def. 56.1 ¶ 27; PI. 56.1 ¶ 27. Walzak wrote in her report that she found “massive volume of material breaches, ranging from serious instances of fraud to Flags-tar’s multiple failures to adhere to its underwriting guidelines and standard industry practices” in both samples of loans. Rendon Deck, Ex. BB, Origination Rep. of Rebecca B. Walzak (“Walzak Origination Rep.”), at 2.
In her report, Walzak also found that Flagstar failed to properly service 128 loans in the 2005-1 sample, or thirty-two percent of the sample, and 201 loans in the 2006-2 sample, or fifty percent of the sample. PI. Reply 56.1 ¶ 21; Def. Reply 56.1 ¶ 21.
*601Over 200 million dollars of loans from the Transactions have been charged off as uncollectable; approximately 71.2 million dollars in the 2005 deal, Buchdahl Decl., Ex. J at 3, and approximately 132 million dollars in the 2006 deal, Buchdahl Decl., Ex. K at 3. Assured has paid more than $90 million in claims. Rendon Decl., Ex. RR at 13.
Under Federal Rule of Civil Procedure 56(a), a party seeking summary judgment must demonstrate that there is “no genuine dispute as to any material fact” and that the party is “entitled to a judgment as a matter of law.” Fed.R.Civ. P.56(a); see also Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). Under New York law, which governs the contractual agreements here in issue, “a written agreement that is complete, clear and unambiguous on its face must be enforced according to the plain meaning of its terms[.]” UBS Financial Services, Inc. v. West Virginia University Hospitals, Inc., 660 F.3d 643, 649 (2d Cir.2011). Where a contract is ambiguous, however, the matter “should be submitted to the trier of fact.” Consarc Corp. v. Marine Midland Bank N.A., 996 F.2d 568, 573 (2d Cir.1993). An insurance policy is no different from a standard contract in that a “provision in an insurance policy is ambiguous when it is reasonably susceptible to more than one reading.” United Air Lines, Inc. v. Ins. Co. of State of Pa., 439 F.3d 128, 134 (2d Cir.2006) (citation omitted).
Flagstar’s primary argument in support of its summary judgment motion is that Assured failed to prove that the breaches caused Assured any actual loss. Of course, causation is ordinarily an “essential element of damages in a breach of contract action.” Nat'l Market Share, Inc. v. Sterling Nat’l Bank, 392 F.3d 520, 525 (2d Cir.2004) (emphasis in original). But, as any first year law student knows, “causation” is far from a self-defining term, and raises all sorts of questions, such as whether the causation must be direct or indirect, transactional, proximate, risk-related, or whatever. Here, moreover, the damages being sought derive in substantial part from defendants’ alleged failure to repurchase the loans, a contractual remedy that was not tied to plaintiffs suffering any damages from the alleged breach.
The circumstances here are nearly identical to those in Syncora Guarantee Inc. v. EMC Mortg. Corp:, 874 F.Supp.2d 328 (S.D.N.Y.2012). As in the instant case, Syncora involved allegations that the defendant (a loan originator) breached representations and warranties made to secure insurance from the plaintiff (a note insurer) in the context of a loan transaction involving HELOCs. In Syncora, the loan documents stated that the repurchase provision was triggered only when a breach “materially and adversely affects the value of the interests of the Purchaser, the Noteholders, the Indenture Trustee or the Note Insurer in any of the HELOCS,” Syncora, 874 F.Supp.2d at 334, and in this case, the loan documents state that the repurchase provision was triggered when a breach “materially and adversely affects the interest of the Issuer, the Noteholders or the Note Insurer in the related Mortgage Loan,” SSA § 2.04(b).
In Syncora, as in the instant case, the defendant argued that the contract required the plaintiff to “show actual pecuniary loss resulting from a breach.” Syncora, 874 F.Supp.2d at 334. Judge Crotty, however, granted the plaintiff note insurer’s motion for summary judgment, concluding that the contracts did not require the plaintiff to show that the breaches caused the loans to default, but only that the breaches “materially increased” plaintiffs risk of loss. Id. at 340. As explained further below, the Court agrees with Judge Crotty’s analysis.
*602Defendants are not liable for breaches unless they are both material and adverse. Therefore, it is necessary to determine what those words mean in the context of the Transaction Documents. According to their dictionary definitions, “material” means “[o]f such a nature that knowledge of the item would affect a person’s decision-making; significant; essential,” Black’s Law Dictionary (7th ed.1999), and “adverse” means “opposed to one’s interests.” Merriam-Webster’s Collegiate Dictionary 19 (11th ed.2003).
Defendants, however, while largely agreeing with the dictionary definition of “material” (they would define “material” as “important to Assured’s decision to insure the Transaction”), propose that “adverse” be here defined as “causes [Assured] actual loss.” Memorandum of Law in Support of Defendants’ Motion for Summary Judgment (“Def. Mem.”) at 15. But defendants offer no good reason for such a substantial departure from the ordinary meaning of “adverse,” and the Court knows of none. Accordingly, the Court gives “adverse” its ordinary meaning of “opposed to one’s interests.” In this context, a breach of contract that materially increased Assured’s risk of loss would be adverse, because it was opposed to the insurer’s interests.
This fully accords with New York law. As Judge Crotty noted in Syncora, “New York law [which governs here] provides that an insurer has an interest in receiving complete and accurate information before deciding whether to issue a policy.” 874 F.Supp.2d at 334. Moreover, New York law recognizes that an insurer may rescind a policy where an insurer has relied on a material misrepresentation, and where “knowledge ... of the facts misrepresented would have led to a refusal by the insurer to make such contract.” Id. (quoting Mut. Benefit Life Ins. Co. v. JMR Elecs. Corp., 848 F.2d 30, 32 (2d Cir.1988) (per curiam)). While this is not a rescission case, plaintiffs immediate contractual remedy — repurchase—is closely akin to rescission, and it is defendants’ alleged refusal to repurchase that largely “caused” the damages here sought.
Furthermore, New York Insurance Law defines warranty as “any provision of an insurance contract which has the effect of requiring ... the existence of a fact which tends to diminish, or the non-existence of a fact which tends to increase, the risk of the occurrence of any loss, damage, or injury within the coverage of the contract.” N.Y. Ins. L. § 3106(a) (emphasis added). That law states that “[a] breach of warranty shall not avoid an insurance contract or defeat recovery thereunder unless such breach materially increases the risk of loss, damage or injury within the coverage of the contract.” Id. § 3106(b) (emphasis added). While it is true that the definition of breach in these laws expressly applies to claims seeking recovery against an insurance company, there is no reason to think that the same definition would not apply to the instant matter, where the insurance company js suing the seller for breach of warranty.
Further still, the SSAs, as an alternative to the remedy of repurchase, contain cure provisions, and if a breach only occurred after the loan had already defaulted, the cure provision would have no meaning. A contract should not be interpreted so as to render a clause superfluous or meaningless. Galli v. Metz, 973 F.2d 145, 149 (2d Cir.1992).
It should also be noted that the Transaction Documents do not mention “cause,” “loss” or “default” with respect to the defendants’ repurchase obligations. If the sophisticated parties had intended that the plaintiff be required to show direct loss causation, they could have included that in the contract, but they did not do so,4 and *603the Court will not include that language now “under the guise of interpreting the writing.” See Syncora, 874 F.Supp.2d at 335 (quoting Reiss v. Fin. Performance Corp., 97 N.Y.2d 195, 199, 738 N.Y.S.2d 658, 764 N.E.2d 958 (2001)); see also MBIA Ins. Corp. v. Countrywide Home Loans, Inc., 34 Misc.3d 895, 936 N.Y.S.2d 513, 522 (N.Y.Sup.Ct.2012) (materiality of breach depends on the “risk of loss”).
Accordingly, for all the foregoing reasons, the Court concludes that plaintiff must only show that the breaches materially increased its risk of loss. Put another way, the causation that must here be shown is that the alleged breaches caused plaintiff to incur an increased risk of loss.
In conjunction with their causation argument above, defendants also argue that the relevant time period for examining breach is within twelve months of the date of origination. Because the Court has already ruled that Assured is not required to prove loss causation in the sense of loan defaults, it is therefore irrelevant for summary judgment purposes whether the loans defaulted within twelve months.
Flagstar next argues that Assured conducted extensive due diligence before the Transactions closed and that it should be bound by what Assured classified as material breaches before the Transactions took place. Before closing on the Transactions, Assured evaluated each of the Transactions to determine whether to provide insurance for them. Def. 56.1 ¶ 10; PI. 56.1 ¶ 10. The Clayton Group (“Clayton”) performed reviews of the 2005-1 deal, and the Bohan Group (“Bohan”) performed due diligence on the 2006-2 deal. Def. 56.1 ¶ IS; PI. 56.1 ¶ 13. Assured directed Clayton and Bohan to review samples of the loans in each of the loan pools to determine whether the loans were in compliance with both Flagstar’s and Assured’s underwriting guidelines. See Def. 56.1 ¶¶ 14,17; PI. 56.1 ¶¶ 14,17.
The reviewers set up a coding system for grading the loans: loans were classified as Event Level 1, Event Level 2, or Event Level 3, with Event Level 3 being the most serious. Def. 56.1 ¶ 18 PI. 56.1 ¶ 18. Assured employee George Stiehl testified that if there were a large number of Event Level 3’s, Assured would have considered not insuring the deal at all. Def. 56.1 ¶ PI. 56.1 ¶ 19. According to Clayton, there were zero Event Level 3’s in the 2005-1 transaction sample it tested. Def. 56.1 ¶ 23; PI. 56.1 ¶ 23. For the 2006-2 Transaction, Bohan reviewed a 250-loan sample and found six Event Level 3’s, but Assured believed that all of these problems would be cured by the time the transaction closed. Def. 56.1 ¶ 25; PI. 56.1 ¶25. There are nineteen loans present in the loan samples and in the due diligence samples. Def. 56.1 ¶ 52; PI. 56.1 ¶ 52. Plaintiffs expert found breaches in seventeen of those nineteen loans; but Clayton and Bohan rated each of the nineteen breaches as Event Level 1 at the time of closing. Def. 56.1 ¶ 53; PI. 56.1 ¶ 53.
However, the third party due diligence firms were not asked to review the loans to determine if they complied with all of the representations and warranties in the Transaction Documents. Buchdahl Deck, Ex. T, Deck of George Stiehl ¶¶ 7, 16. Moreover, the third-party reviews did not look for evidence of fraud. Buchdahl Deck, Ex. U, Deposition of Vicki Beal, Dec. 14, 2011 (“Beal Dep.”) at 138:6-15. Viewing the facts in the light most favorable to Assured, Assured obtained the rep*604resentations and warranties to ensure the quality of the loans in part because its reviews were not comprehensive. Moreover, the representations and warranties do not mention Assured’s due diligence. Assured obtained those representations and warranties even after conducting its own diligence review. Under defendants’ theory, the plaintiff would be prohibited from claiming breach once it entered into the Transactions, because the plaintiff had hired firms to conduct a due diligence review and they had found that there were no Event Level 3 breaches. But this would render the representations and warranties in the Transaction Documents entirely superfluous, and the Court cannot endorse such a reading. At this stage, there is at least a genuine issue of material fact as to whether the due diligence reviews defined materiality for the life of the loans, and the Court is skeptical that they did.
If plaintiff is not limited to its pre-transaction definitions of materiality, Flagstar next argues that plaintiff has waived its right to recover for any breaches based on the conclusions of the due diligence review. Def. Mem. at 21. But in CBS Inc. v. ZiffDavis Publishing Co., 75 N.Y.2d 496, 554 N.Y.S.2d 449, 553 N.E.2d 997 (1990), the New York Court of Appeals held that the a buyer can recover for breach of warranty even if the purchaser had formed doubts as to the truth of the warranted facts prior to the closing. To hold otherwise, “would have the effect of denying the express warranties of their only value to [the purchaser] — i.e., as continuing promises by [the seller] to indemnify [the purchaser] if the facts warranted proved to be untrue.” Id. at 506, 554 N.Y.S.2d 449, 553 N.E.2d 997. The Court of Appeals adopted the “basis of the bargain” approach to this issue, holding that “[t]he critical question is not whether the buyer believed in the truth of the warranted information .... but whether [the buyer] believed [the buyer] was purchasing the [seller’s] promise [as to its truth].” Id. at 503, 554 N.Y.S.2d 449, 553 N.E.2d 997.
The exception to the Ziff-Davis rule is if a “buyer closes on a contract in full knowledge and acceptance of facts disclosed by the seller which would constitute a breach of warranty.” Galli v. Metz, 973 F.2d 145, 151 (2d Cir.1992). Defendants argue that they provided detailed loan information to plaintiff and that plaintiff nonetheless agreed to insure the Transactions. The Second Circuit has made clear that in order to meet this exception, however, the critical question is not what the buyer knew, but “whether he got that knowledge from the seller.” Rogath v. Siebenmann, 129 F.3d 261, 265 (2d Cir.1997). If the buyer “has been informed of the falsity of the facts by some third party,” he has not waived the representations and warranties. Id. Defendants argue that they provided the information to plaintiff, but their argument stretches the exception too far. Here, although the data came from Flagstar, the relevant information came from third party due-diligence firms who were retained by JP Morgan, another third party. Flagstar cannot successfully argue that it disclosed the relevant facts to Assured; under Flagstar’s theory, a seller could always just dump a mountain of data on a buyer and then if the buyer had other firms evaluate the data and still went through with the transaction, the seller could argue that any express warranties had been waived. Moreover, unlike in Galli, the parties here did not agree that any of the representations and warranties were inaccurate. Therefore, Assured has not waived its right to assert breaches here.
Flagstar next argues that Assured is responsible for any losses it suffered because its loss model was “intentionally optimistic.” Def. Mem. at 28. To design an *605adequate cushion to protect against unexpected loss, Assured engaged in loss modeling. Def. 56.1 ¶ 35; PI. 56.1 ¶35. Assured used a loss model that took the average of three separate loss models: the “RMG Model,”5 the “FICO Model,” and the “Historic Model.” Def. 56.1 ¶ 36; PL 56.1 ¶ 36. For the 2005-1 Transaction, the FICO Model produced an expected loss of 1.28%, the Historical Model produced an expected loss of 1.74%, and the RMG model produced an expected loss of 9.94%. Rendon Decl., Ex. F at 6. When these three figures were averaged, the total expected loss figure was 4.32%. Id. For the 2006-2 Transaction, the FICO Model produced an expected loss of 1.10%, the Historical Model produced an expected loss of 1.74%, and the RMG Model produced an expected loss of 8.05%. Def. 56.1 ¶ 38; Pl. 56.1 ¶38. By taking the average of the three models, Assured calculated an expected loss for the 2006-2 Transaction of 3.63%. Rendon Decl., Ex. G at 5. Assured likely would not have done the deal if the loss ratios had been around 9 or 10 percent. Def. 56.1 ¶ 41; Pl. 56.1 ¶41. As discussed above, plaintiffs loss model was an average of three loss-models, which resulted in a much lower rate of projected loss than plaintiffs proprietary RMG loss model.
Although the average of the three models produced a lower expected loss than the RMG model alone, David Beard, the Assured employee who was responsible for the Flagstar deal, testified that the proprietary RMG loss model had an “upward bias” because it was designed for subprime loans. Buchdahl Decl., Ex. Y, Deposition of David Beard, Nov. 1, 2011 (“Beard Dep.”) at 189:8-16. Beard testified that if the Assured team felt that averaging were not appropriate, the team would not have used averaging. Id. at 195:2-3. Mr. Beard’s supervisor also testified that he still believed that the averaging was “appropriate.” Buchdahl Decl., Ex. Z, Deposition of David Williams, Nov. 8, 2011 (“Williams Dep.”) at 227:8-11.
None of this provides a basis for granting summary judgment, for several reasons. First, at least at this stage, plaintiffs due diligence cannot override the express warranties and representations provided by the defendants. Moreover, even assuming arguendo that the loss model was relevant, there is at least a genuine issue of material fact as to the propriety of plaintiffs use of an average model. It may well be that plaintiff should have simply used its proprietary RMG model, and in hindsight, it would have been better off it had used that model. But plaintiff has put forth evidence that the single model may have been too pessimistic because it was designed for sub-prime loans, and resolution of this factual dispute is not appropriate at this stage.
Flagstar also argues that its expert has concluded that the plaintiff will not suffer any losses for the 2005-1 Transaction based on the predicted cash flows of that Transaction. This argument is only relevant to the 2005-1 Transaction and does not affect the 2006-2 Transaction. Although Assured has paid millions of dollars in claims on this Transaction so far, defendants assert the 2005-1 Transaction is still receiving payments and that in the end, there will be enough money in the 2005-1 trust to repay Assured for any claims payments it made.
Flagstar’s residual interest in the Transactions is reflected in the “Transfer- or’s Interest.” The Transferor’s Interest in the 2005-1 Transaction is currently $11 million, and Flagstar asserts that Assured *606therefore cannot suffer any damages on the 2005-1 Transaction. But Flagstar employee Stanley Jursek testified that the original modeling of the Transferor Interest was “[m]ost likely” more than $20 million, almost twice as much as the $11 million that Flagstar now asserts is available as Transferor’s Interest. See Buchdahl Deck, Ex. C, Deposition of Stanley Jursek, Oct. 28, 2011 (“Jursek Dep.”) at 106:20-21. Therefore, the amount of Transferor Interest is not readily susceptible to determination on summary judgment, and the Court cannot grant summary judgment on this basis. There is no provision in the contract that requires Assured to wait until all activity on the transactions is complete before obtaining any recovery. Moreover, if the fund does recover more in payments than Assured has predicted, Assured will be responsible for paying this money to Flagstar.
Defendants also argue that plaintiffs method of extrapolating from breaches in the sample to the rest of the loans in the transaction does not identify which loans contain breaches and “deprives Flagstar of its bargained for right to cure such alleged breaches.” Def. Mem. at 80. But the cure or repurchase period ends 90 days after defendants learned of the loan breaches, see SSA § 2.04(d), and that time period has long since expired.6
Finally, the Court turns to the servicing claims. Under the terms of the contract, defendants argue that their liability for servicing violations is limited to bad faith and grossly negligent conduct. See SSA § 5.03.
Section 5.03 of the SSA provides:
The Servicer and each of its directors, officers, employees and agents shall be indemnified by the Trust and held harmless against any loss, liability, or expense incurred in connection with any legal action relating” to its obligation as a services” unless inter alia such “loss, liability, or expense is due to its willful misfeasance, bad faith or gross negligence in the performance of its duties ... or due to its reckless disregard of its obligations.
Id.
Plaintiff responds that this provision only applies to claims brought by the “Issuer, the Owner Trustee, the Transfer- or, or the Noteholders.” See id. It does not, however, explicitly apply to the Note Issuer, Assured. While this is true, this provision applies equally to Assured because it is a third-party beneficiary to the SSA and is thus bound by its terms. As a third-party beneficiary, Assured “possessed no greater right to enforce a contract than the actual parties to the contract.” BAII Banking Corp. v. UPG, Inc., 985 F.2d 685, 697 (2d Cir.1993).
The primary evidence that plaintiff offered in support of its servicing claim was the report by its mortgage servicing expert, Rebecca Walzak. Walzak testified, however, that she was not asked to consider the gross negligence, misfeasance, or bad faith standard established by the SSA.7 Even though it is permissible for *607an expert to testify to ultimate issues, the expert need not attempt to make the ultimate legal determination as to whether the defendant was hable under the gross negligence standard. Ultimately, the trier of fact will make that determination, and the question on summary judgment is whether, viewing the facts in the light most favorable to Assured, there is a genuine dispute of material facts on this issue. The Court concludes that there is a genuine dispute.
The SSAs specifically state that Flags-tar “shall service and administer the Mortgage Loans in a manner consistent with the terms of this Agreement and with ‘Customary Servicing Practices.’ ” SSA § 3.01. Walzak testified that she calculated breaches of the servicing standard based on a failure to comply with the Customary Servicing Practices required by the SSA. She further declared that her evaluation of Customary Servicing Practices was derived from standards set forth by Fannie Mae, Freddie Mac, and HUD. Buchdahl Deck, Ex. EE, Deposition of Rebecca Walzak, Dec. 19, 2011 (‘Walzak Dep.”) at 80:8-15. Walzak concluded that Flagstar failed to service 32 percent of the loans in the 2005-1 random sample and 50.2 percent of the loans in the 2006-2 random sample in compliance with Customary Servicing Practices as required by the SSAs. Buchdahl Deck, Ex. DD, Corrected Supp. Expert Rpt. of Rebecca B. Walzak, (Walzak Servicing Rpt.”) at 2. At trial, the trier of fact could conclude that these widespread failures to follow customary servicing practices constituted “gross negligence, misfeasance, or bad faith.” Although the Court is skeptical that plaintiff can meet this standard at trial, the Court finds that the plaintiff has presented enough evidence to survive summary judgment on this claim.
Defendants also dispute Walzak’s methodology and coding system for the servicing problems she identified, but the Court finds that these issues are inappropriate for resolution at this stage. For example, defendants argue that almost all of the issues that Walzak categorized as servicing problems were labeled as Code 2, which Walzak stated were issues that were “not significant enough that I thought there were going to be further complications or problems” or that something else in the file had “remediated [those issues] in some fashion or another.” Walzak Dep. at 167:10-15. There is a genuine dispute of material fact about the severity of these violations. For example, in her written report, Walzak explained that 226 Code 2 loans were labeled as such because Flags-tar had “failed to obtain an independently derived market value for the Mortgage Property.” Walzak Servicing Rpt. at 14.
Defendants then proceed to nitpick through the rest of Walzak’s identified problems in a manner that is inappropriate at this stage. For example, they argue that six of the twenty assets marked as Code 3 were so marked because they contained multiple Code 2 errors and thus are not serious enough to consider. Defendants seek to undercut the remaining fourteen Code 3 violations by stating that four of the errors “involve the nature or position of the loans themselves” and that the remaining ten errors involved “judgment calls,” Def. Mem. at 34, which could be no more than “errors in judgment.” These disputed factual arguments are inappropriate for resolution on summary judgment, and instead they are properly reserved for the trier of fact.
The Court has considered defendants’ additional arguments and finds them without merit. Therefore, for the foregoing reasons, the Court reaffirms its bottom-line order denying defendants’ motion for summary judgment. Trial of this matter *608will therefore commence, as scheduled, on October 9, 2012.
. Assured was formerly known as Financial Security Assurance, Inc., or “FSA.” For the sake of consistency, this Memorandum will refer to this entity as "Assured” throughout.
. The underlying collateral for the securities was actually second-lien home equity lines of credit ("HELOCs”). This opinion will follow the parties' lead and refer to these HELOCs simply as “loans.”
. As the Court noted in its opinion on Flags-tar’s motion to dismiss, the "contractual terms of the 2005 and 2006 versions of these [six] documents are 'materially identical’ to one another.” Assured Guar. Mun. Corp. v. Flagstar Bank, FSB, 11 CIV. 2375, 2011 WL 5335566, at *1 n. 1 (S.D.N.Y. Oct. 31, 2011).
. Indeed, in Flagstar's 2006-2010 annual reports, in a discussion of its repurchase obli*603gations, Flagstar stated that those obligations extended to loans that were currently performing. Buchdahl Decl., Ex. N at 44 ("Loans that are repurchased and that are performing according to their terms are included within our loans held for investment portfolio.”)
. The RMG Model was the only model that was based on Assured’s loan file review of the loans to be included in the 2005-1 and 2006-2 Transactions. Def. 56.1 ¶ 43; Pl. 56.1 ¶ 43.
. Flagstar also raises challenges to plaintiffs experts. Most of those challenges are disposed of by the Court’s ruling on the causation matter discussed above. The other challenges are not suitable for resolution at this stage.
. Gross negligence is "conduct that evinces a reckless disregard for the rights of others or 'smacks’ of intentional wrongdoing.” AT & T v. City of New York, 83 F.3d 549, 556 (2d Cir.1996) (quoting Colnaghi, U.S.A., Ltd. v. Jewelers Protection Servs., Ltd., 81 N.Y.2d 821, 823-24, 595 N.Y.S.2d 381, 611 N.E.2d 282 (1993)).