The US Treasury recovers billions of dollars thanks to AI

Related

How Bitcoin’s price rise has increased the number of cryptocurrency payments 

NOWPayments Announces Significant Gain in Crypto Payments NOWPayments, a leading...

Bitcoin ETF spot: weekly inflows of 1.7 billion dollars

The spot ETFs on Bitcoin continue to register an...

Mike Tyson: in addition to the passion for boxing, there is also one for Bitcoin and crypto

Mike Tyson, famous boxer and 58-year-old entrepreneur, has repeatedly...

Metaplanet issues 1-year bonds to purchase new Bitcoin (BTC)

Metaplanet has announced its intention to issue 1-year bonds...

The class action Dogecoin against Elon Musk concludes as the market takes off: all the details

After the accusations of market manipulation by Dogecoin investors,...

Share

The US Department of the Treasury in the fiscal year 2024 recovered billions of dollars in fraud using artificial intelligence (AI)-based machine learning to analyze data and identify fraud.

The same Department of the Treasury revealed it yesterday with an announcement on their official website. 

The use of AI by the US Treasury

The sparse statement reveals that the US Treasury has improved its fraud detection processes, using among other things artificial intelligence based on machine learning.

In particular, starting from the pandemic, the Office of Payment Integrity (OPI) began using advanced processes to counter the increase in fraud and improper payments. 

Thanks to these efforts, and the adoption of a technology and data-driven approach, they managed to prevent and recover over 4 billion dollars in fraud and improper payments from October 2023 to September 2024.

In the previous fiscal year, only 652.7 million dollars had been recovered, and the increase is due precisely to the efforts of the OPI to improve its fraud prevention capabilities and expand the offerings to new and existing clients.

The Treasury’s OPI shares

The OPI in this way managed to expand risk-based screening, obtaining 500 million dollars in fraud prevention, and also managed to identify and prioritize high-risk transactions, resulting in a prevention of 2.5 billion dollars.   

Additionally, they have implemented efficiencies in the payment processing program, resulting in the prevention of 180 million dollars. 

So some of the 4 billion dollars mentioned above have actually been recovered, while the majority comes from prevention, meaning funds that would have ended up in the hands of fraudsters if they had not been identified and stopped. 

A billion dollars has been effectively recovered thanks to the acceleration in the identification of fraud with intelligenza artificiale based on machine learning.

The US Treasury has also stated that it is focusing on creating and strengthening partnerships with new and high-risk programs, to increase access to and use of payment integrity solutions, such as state-administered programs funded at the federal level. 

In this regard, in May of this year, together with the Department of Labor, they announced a data-sharing partnership to provide state unemployment agencies with access to data sources and services of the Do Not Pay Working System through the Unemployment Insurance Integrity Data Hub. 

The utility of artificial intelligence

Often, scammers use well-tested procedures and schemes to try to get away with it. 

With artificial intelligence, it is possible to quickly examine enormous amounts of data in search of signs of those possible patterns. In this way, not only is it possible to greatly speed up the identification of suspicious transactions, but it is also possible to examine more of them. 

Furthermore, it can also be used to better evaluate, in an objective manner, the risk that a given transaction may be improper or fraudulent, thus improving the effectiveness of fraud detection work. 

Machine learning furthermore allows for the continuous improvement of the effectiveness of these investigations, thus making life increasingly difficult for the truffatori. 

Also because scammers are becoming increasingly skilled at hiding within the crowd, especially when the latter is very large. 

The Impressive Numbers of the US Treasury: Will AI Be Helpful?

The US Treasury states that it disburses as much as 1.4 billion payments per year to more than 100 million people, for a total value of over 6.9 trillion dollars. 

These numbers are so high that the 4 billion dollars mentioned above pale in comparison. 

In the official statement, they also specify that at this moment, losses due to fraud in the financial sector continue to increase every year, with online payment fraud expected to cumulatively exceed 362 billion dollars by 2028.

The Department of the Treasury proactively supports federal programs to mitigate the risks of financial fraud by leveraging data and emerging technologies.

“`html

The comment

“`

The Deputy Secretary of the Treasury, Wally Adeyemo, stated: 

“The Treasury takes our responsibility seriously to act as effective stewards of taxpayer money. Helping to ensure that agencies pay the right person, the right amount, at the right time is crucial to our efforts. We have made significant progress over the past year in preventing over 4 billion dollars in fraudulent and improper payments. We will continue to collaborate with others in the federal government to equip them with the tools, data, and expertise necessary to stop improper payments and fraud”.