The latest benefit from the artificial intelligence revolution: saving American taxpayers billions of dollars.

The U.S. Treasury Department said its increased use of artificial intelligence in fraud detection helped prevent or recover more than $4 billion over the past year.

In a release Thursday, Treasury officials said it had been leaning especially into machine learning, which involves analyzing large quantities of data, to detect patterns of fraud commonly used by criminals.

The results:

"Treasury takes seriously our responsibility to serve as effective stewards of taxpayer money, Deputy Secretary of the Treasury Wally Adeyemo said in a statement. "Helping ensure that agencies pay the right person, in the right amount, at the right time is central to our efforts."

A Treasury official declined to provide specific examples of how AI had been deployed, given the nature of the activities targeted. The Department said it plans to lend resources and lessons-learned from its AI efforts to other federal agencies.

Online payment fraud is set to cumulatively surpass $362 billion by 2028, according to UK-based consultancy Juniper Research. Already, check fraud in the U.S. has surged by 385% since the pandemic, according to Treasury.

Check fraud, whether involving government or privately issued funds, continues to proliferate, with an estimated $688 million in reported suspicious activity filed by chartered banks between February and August 2023.

Treasury joins the Internal Revenue Service in tapping into improved technology to create fiscal gains. Last year, the IRS announced it had begun increasing deployment of AI to better detect tax cheats, identify compliance threats, and improve case selection tools "to avoid burdening taxpayers with needless 'no-change' audits."