WASHINGTON, D.C. — In a Washington AI Network podcast interview with host Tammy Haddad, cybersecurity veteran and author of “Dark Side of the Boom: How Scammers Hijack the AI Revolution” Patrick Coughlin examines how artificial intelligence has turned digital fraud into America’s fastest-growing — and now largest — economic crime and urges federal leadership, international cooperation, and know-your-customer rules for AI model access.
Coughlin co-founded Savi Security with his brother, Ryan, after their mother, Beth Coughlin, then 73 and living alone in Prairie Village, Kansas, was targeted by an AI-generated voice-cloning “kidnapping” scam. Scammers used a cloned voice that convincingly mimicked Coughlin’s sister to tell their mother she had been kidnapped, then demanded $1,200 for her release. Although Beth Coughlin ended the call and confirmed her daughter was safe before sending any money, the experience left her badly shaken — and sent Patrick Coughlin, a two-decade veteran of cybersecurity, intelligence, and emerging technology, into what became years of research into global scam syndicates.
That research became the basis for “Dark Side of the Boom,” and ultimately for Savi Security, which Coughlin and his brother founded to build consumer-focused protection against AI-enabled fraud. The company raised $7 million this year and launched a consumer app designed to help everyday people identify scams before they become victims.
A Crime Growing Faster Than Any Regulatory Response
Citing data from the FTC and FBI, Coughlin told Haddad that Americans lost an estimated $200 billion to digital fraud over the past 12 months, with losses growing 25–30% year over year — a pace he compared to the revenue growth of top AI chip companies. Digital fraud, he said, is now both the fastest-growing and the largest economic crime in the United States.
Coughlin argued that the public conversation about AI has focused disproportionately on enterprise and national-security use cases, leaving everyday consumers out of the debate. “If you go to Middle America today, AI is a bad word,” he said. “Their experience with AI is in the fearmongering headlines and in the increase in scam and spam calls, messages, and emails they’re all getting.”
A Policy Agenda for Washington
Coughlin outlined several concrete steps he believes federal policymakers should take, including federal leadership over the current patchwork of roughly 1,200 state-level AI bills, which he said creates gaps and loopholes that scammers can exploit as they move across state lines.
Coughlin also made the case for international cooperation, since the majority of scam operations that target American consumers originate overseas, including in Southeast Asia, Eastern Europe, and West Africa. He also advocated for a know-your-customer approach for access to sensitive components of AI models, such as bulk API access, similar to existing controls on bulk fertilizer or pharmacy purchases. Coughlin argued for enforcement of existing anti-fraud and anti-money-laundering laws against platforms that knowingly profit from fraudulent activity. He pointed to Pulitzer Prize-winning Reuters reporting that found roughly 10% of Meta’s ad revenue comes from fraudulent advertisers — activity he said the company manages as a revenue stream rather than eliminates. “If it was covered by the existing mandates, then you’d see enforcement actions against the companies who we know are violating them,” he said.
Coughlin pushed back on the argument that AI regulation would put the United States at a disadvantage against China, noting that China has implemented far more extensive AI governance, including mandatory watermarking. “If these models are half as dangerous as the people creating them are constantly telling us they are,” he said, “then don’t you think that we should create some more regulation around who gets to access them?”
About the Book
“Dark Side of the Boom: How Scammers Hijack the AI Revolution,” by Patrick Coughlin, examines how AI has industrialized scams and fraud against everyday consumers and calls for a renewed focus on mitigating the real, present-day harms of AI alongside its benefits.