Mortgage Lenders Push for Direct IRS Data Access to Combat AI-Driven Fraud
Financial institutions are increasingly advocating for direct, real-time access to Internal Revenue Service (IRS) income verification data to mitigate the rising risk of AI-generated mortgage fraud. During recent Senate Banking Committee hearings, industry representatives argued that current verification processes are vulnerable to sophisticated digital forgeries, as synthetic documents created by generative AI can easily bypass traditional manual reviews.
The Shift Toward Automated Income Verification
The primary concern for lenders is the precision of AI in creating fraudulent tax documentation. According to the Financial Crimes Enforcement Network (FinCEN), the integration of generative AI tools has lowered the barrier to entry for financial fraud, allowing bad actors to produce high-fidelity counterfeit W-2s and tax returns.
Lenders currently rely on third-party aggregators or manual document uploads to verify borrower income. However, these methods often fail to detect “synthetic identities”—profiles built using a mix of real and fake information. By gaining direct access to the IRS Income Verification Express Service (IVES), lenders aim to verify tax data directly at the source, effectively eliminating the possibility of relying on falsified physical or digital documents.
Regulatory Hurdles and Privacy Concerns
Expanding access to taxpayer data remains a complex regulatory challenge. The Internal Revenue Code Section 6103 strictly limits the disclosure of tax return information to protect borrower privacy. While the IRS already provides a mechanism for third-party verification through IVES, the process is often criticized by lenders for its latency.
Industry participants have noted that while the current system works for standard loan underwriting, it is not optimized for the speed required in modern digital lending. Advocates for the change argue that a modernized API-based connection to the IRS could provide instantaneous verification, reducing the “window of opportunity” for fraudsters to submit altered documents during the loan origination phase.
Impact on Mortgage Underwriting Standards
The push for improved data access represents a broader transition in how lenders assess credit risk. As AI becomes more prevalent in both the origination and the detection of fraud, the mortgage industry is moving away from document-based verification toward data-centric models.
| Verification Method | Reliability | Speed |
| :— | :— | :— |
| Manual Review | Low | Slow |
| Third-Party Aggregators | Moderate | Fast |
| Direct IRS (IVES) API | High | Real-time |
By shifting to direct data verification, lenders expect to reduce the volume of “buyback” requests—where secondary market investors force lenders to repurchase loans found to contain inaccurate or fraudulent borrower data.
Looking Ahead: Next Steps for Financial Oversight
The debate over IRS data access is expected to continue as lawmakers balance the need for fraud prevention with the protection of taxpayer privacy. Future legislative discussions will likely focus on whether to grant broader access to non-bank mortgage lenders or fintech platforms that currently operate under different regulatory frameworks than traditional depository institutions.
For now, the mortgage sector continues to enhance its internal AI-detection software to flag inconsistencies in submitted documents. However, industry leaders maintain that until the verification process moves to a source-of-truth model—where lenders verify data directly with the government agency holding the records—the risk of AI-manufactured deception remains a primary threat to market integrity.
Related reading