Global fraudsters are using artificial intelligence and polished social engineering to create more realistic scams that target individuals and organizations [1, 2].

This evolution in fraud represents a critical shift in cybersecurity, as AI-powered tools allow scammers to bypass traditional red flags and personalize attacks at scale. The ability to mimic official government agencies makes these threats harder for the average citizen to detect.

Law enforcement agencies, including the King County Sheriff’s Office, said that scams are becoming more sophisticated [2]. In the U.S., fraudsters have posed as government entities such as the Nevada Department of Motor Vehicles to deceive victims [3]. Local victims in some areas have lost thousands of dollars to these advanced phone scams [2].

Experts said this surge is due to the adoption of AI-powered malware and generative tools that make fraudulent messages appear authentic [3, 4]. These tools allow scammers to refine their language and tactics, removing the grammatical errors and awkward phrasing that previously served as warning signs for victims [4, 5].

Eva Velasquez, CEO of the Identity Theft Resource Center, said, "A major factor behind that increase is the growing sophistication of scams, such as the use of artificial intelligence" [3].

Governments and private companies have begun coordinating "fight-back" efforts to counter these trends [1]. These initiatives focus on developing new defensive measures, and increasing public awareness about the capabilities of AI-driven fraud. The surge in these activities has been particularly noticeable throughout 2025 and 2026 [3, 4].

“AI-powered tools allow scammers to bypass traditional red flags and personalize attacks at scale.”

The integration of AI into the scam industry lowers the barrier to entry for high-quality social engineering. As fraudulent communications become indistinguishable from legitimate government or corporate outreach, the burden of verification shifts from identifying 'fake' markers to implementing systemic, multi-factor authentication and zero-trust protocols.