Treasury says Do Not Pay now reaches 99% of federal programs, with $3.7 trillion in payments screened
The US Treasury reported its Fiscal Year 2026 results on stopping fraud and improper payments. Access to its Do Not Pay screening tools rose from about 4 percent of federal programs to about 99 percent.
On October 6, 2026, the US Department of the Treasury and the Bureau of the Fiscal Service announced their Fiscal Year 2026 results on preventing fraud and improper payments. Treasury says the work fulfils key requirements of Executive Order 14249, Protecting America's Bank Account Against Fraud, Waste, and Abuse, issued on March 25, 2025.
Do Not Pay is a Treasury program that gives agencies data and tools to check identity and eligibility before money goes out. Treasury says about 99 percent of federal programs can now access all Do Not Pay data sources they are allowed to use, compared with about 4 percent at the end of FY2025. Most remaining programs are on track to finish onboarding in early FY2027. Treasury screened over 2.3 billion records against Do Not Pay data sources, up from 641 million records in FY2025, and added nine new datasets, including company registration information through OpenCorporates.
Treasury also introduced a new payment verification process in FY2026. Under it, Treasury screened more than 1.1 billion federal payments totaling about $3.7 trillion. It identified and returned about 13,500 payments, totaling $175 million, that would have gone to deceased individuals.
Treasury also piloted checks that confirm who owns a bank account and whether Taxpayer Identification Numbers (TINs) are present and correctly formatted on federal payments. Treasury says these checks became fully operational on September 30, 2026.
Why it matters to AML professionals
Government payments are a common target for fraud, and the proceeds often pass through bank accounts that AML professionals monitor. Pre-payment checks such as account ownership and identity verification mirror the controls banks use at onboarding, so weak spots shift toward accounts that pass these checks but are controlled by someone else. Fraud and AML teams can expect fraudsters to adapt by using mule accounts and synthetic or stolen identities.
Red flags in this case
- Government benefit or tax refund deposits into an account whose holder name does not match the named payee
- Multiple government payments for different named individuals landing in one account
- Government payments credited to an account after the account holder has died
- Rapid withdrawal or onward transfer of a government deposit soon after it arrives
If this landed in your queue
- Compare the payee name on the incoming government payment with the account holder name and KYC records
- Check whether the account has received payments for several different individuals or from several benefit programs
- Review account activity after the deposit for fast cash-outs, peer-to-peer transfers or crypto purchases
- Escalate for a SAR review and contact the paying agency through proper channels if the funds appear misdirected
Key terms
- Improper payment
- A government payment that should not have been made or was made in the wrong amount, including payments obtained by fraud.
- Taxpayer Identification Number (TIN)
- A number used by US tax authorities to identify a person or business, such as a Social Security number or employer ID number.
- Account ownership verification
- A check that confirms the bank account receiving money belongs to the intended payee.
Interview angle
A customer's account receives several government benefit payments in different names. How would you investigate?
A strong answer covers checking payee names against KYC data, reviewing what happened to the funds after deposit, looking for links to other accounts, and explaining when to escalate for a SAR.
Sources
Go deeper on AMLift
Every fact in AMLift News Desk stories comes from an official source: a regulator, court, government gazette or the company's own filing.
Report an error
Every report is checked against the official source. If we got something wrong, the story is corrected.