IRS Investigators Report AI-Powered Tools Are Fueling a Surge in Crypto Fraud

Federal investigators are sounding the alarm over a significant increase in sophisticated cryptocurrency fraud schemes, attributing the surge to criminals leveraging artificial intelligence to manipulate victims and obscure illicit transactions. Officials from the Internal Revenue Service's Criminal Investigation (IRS-CI) division recently detailed how AI tools, often available on the dark web, are making scams more effective, efficient, and voluminous, leading to billions in losses for American individuals and businesses. The schemes represent a new evolution in financial crime, combining the anonymity of cryptocurrency with the persuasive power of AI-driven social engineering. One victim, Kyle Holder, lost her entire life savings of $300,000 in under three months after being targeted by scammers who used AI to manipulate her. According to a CBS News report, her case is one of thousands that contributed to an estimated $20 billion in cyber theft in 2025, with more than half of those stolen funds being in cryptocurrency, according to the FBI. Harry Chavis, the special agent in charge of IRS-CI’s New York Field Office, explained that criminals are using dark web AI tools to write targeted scripts and identify potential victims, including those who have been scammed before. "There's other data that has been scraped or pulled through hacks or leaks that can be sold for criminals to obtain. And then they're using these dark AI tools to write scripts to literally go specifically to the victim," Chavis told CBS News. Once a victim sends money, the funds are quickly laundered through a complex web of digital wallets. In Holder's case, her money was consolidated with funds from other victims into wallets that ultimately accumulated millions of dollars. An IRS-CI diagram illustrated how funds from multiple victim wallets are mixed and funneled through a cryptocurrency exchange, which acts as an "off-ramp" for the criminals to convert the digital assets into cash, making the money exceptionally difficult to trace. To combat this growing threat, the IRS is fighting fire with fire. The agency is deploying its own artificial intelligence and machine learning tools to analyze vast datasets and detect criminal activity. Jarod Koopman, IRS-CI's executive director of cyber and forensics, stated that AI is accelerating not only fraud but also the agency's ability to fight it. "AI has certainly made us much more effective and efficient in sifting through that to find patterns, to identify fraud methodologies or typologies, to identify current threats or vectors that might be red flags of issues that we were not aware of prior," Koopman said in a statement. This effort is spearheaded by IRS-CI’s Advanced Analytics and Information (AAI) section, as detailed in the agency's 2024 annual report. The AAI’s primary function is to develop and deploy advanced algorithms to uncover patterns and anomalies indicative of tax fraud, money laundering, and cybercrimes concealed within legitimate financial transactions. This unit integrates machine learning into operations, enabling agents to trace illicit funds and build stronger cases more effectively. The agency's enhanced technological capabilities are already yielding results in the digital asset space. In February 2024, IRS-CI secured the first-ever indictment solely for the underreporting of cryptocurrency gains, charging an individual who allegedly concealed substantial capital gains from the sale of $4 million in bitcoin. This demonstrates an increasing focus and capability in policing the crypto ecosystem, moving beyond just fraud to tax compliance. The agency has also successfully partnered with blockchain intelligence firms to dismantle criminal enterprises like the xDedic darknet marketplace. The technological arms race between criminals and regulators is escalating rapidly. While it's encouraging to see the IRS adopt advanced analytics, our experience shows that businesses cannot afford to be passive observers. Relying solely on law enforcement to prevent or recover losses from these AI-driven scams is a losing strategy. These schemes are designed to exploit human psychology, bypassing technical defenses to trick even cautious employees and executives. A convincing deepfake video call from a supposed CEO or key vendor can be all it takes to authorize a fraudulent transfer. This new reality demands a proactive and layered approach to security. For small and mid-sized companies, this underscores the critical need for robust internal controls and continuous vigilance, which is a core component of effective financial risk management. At C&S Finance Group LLC, we help clients build and test these internal financial safeguards to protect against increasingly sophisticated threats. Business owners can learn more about fortifying their operations by visiting us at csfinancegroup.com. Looking ahead, the dynamic between AI-powered crime and AI-powered enforcement is set to intensify. Businesses and financial professionals should anticipate the continued evolution of these threats and stay informed on guidance from federal agencies, as both criminals and investigators refine their use of artificial intelligence in the ongoing battle over financial security.