---

Artificial intelligence is giving criminals powerful new ways to impersonate relatives, professionals, and trusted institutions, witnesses told the Senate Special Committee on Aging at a hearing focused on the rapidly evolving threat to older Americans.

The bipartisan hearing, titled “The AI Deception Machine: Deepfakes, Chatbots, and the New Frontier of Senior Fraud,” brought together victims, banking and legal experts, a physician targeted by deepfake advertising, and a consumer advocate. It was led by Chairman Rick Scott (R-Fla.) and ranking member Kirsten Gillibrand (D-N.Y.).

“Whether it’s a phone call from someone posing as a grandchild in trouble, a suspicious investment scheme delivered through the mail, or an email from a government imposter threatening jail time, these criminals are targeting seniors with increasing sophistication,” Scott said.

---

Fake kidnapping using a cloned voice

The hearing’s most vivid account came from Deborah Del Mastro, a 64-year-old California woman who said she received a call in May from a man claiming to belong to a Mexican drug cartel. The caller said her daughter had been kidnapped and then played what sounded like the daughter crying and pleading for help.

Convinced the threat was genuine, Del Mastro spent roughly five and a half hours driving among retailers in the San Francisco Bay Area to transfer money to Mexico. She later learned that her daughter was safe and that scammers had apparently cloned her voice. 

The committee also heard from Dr. David Amron, founder and medical director of the Roxbury Institute. Amron said scammers combined footage from his YouTube channel with AI-generated versions of his voice and likeness to produce advertisements that falsely portrayed him as endorsing a “miracle” treatment for lipedema.

Although Dr. Amron reported the advertisements to online platforms, search engines, and the website’s domain registrar, new versions continued to appear after others were removed.

---

A powerful tool for criminals

The accounts illustrated a central concern raised throughout the hearing: AI has not invented the emotional tactics behind fraud, but it has made them easier to deploy at scale. Criminals can now produce personalized messages, realistic voices, fabricated videos, and convincing online personas with limited technical expertise and little expense.

“Generative AI is not replacing traditional scams. It is industrializing them,” Paul Benda, executive vice president for risk, fraud, and cybersecurity at the American Bankers Association, said in prepared testimony. He urged policymakers to preserve the defensive use of AI by financial institutions, which employ automated systems to identify unusual transactions and other signs of exploitation.

Benda also emphasized the role of bank employees, particularly when an older customer appears frightened, secretive, or coached while requesting an unusual transfer. In such cases, he said, personal judgment and an established relationship with the customer can be as important as an algorithm.

---

Rising financial stakes

The financial stakes are rising. A committee report released alongside the hearing said Americans reported nearly $21 billion in cybercrime losses in 2025, including $893 million associated with AI-enabled scams across more than 22,000 complaints. The report also said more than 82% of phishing emails are now produced with AI assistance. 

Witness Matthew Ferraro, a cybersecurity and emerging-technology lawyer, cited an industry projection that generative AI could help drive U.S. fraud losses to $40 billion by 2027, with older people bearing a disproportionate share.

The policy discussion extended beyond consumer education. Witnesses advocated for faster sharing of fraud intelligence, closer federal-state cooperation, stronger authentication and identity systems, greater platform accountability, and more effective enforcement against organized criminal networks.

---

Emerging legislation

Sen. Mark Kelly (D-Ariz.) used the hearing to promote the bipartisan Aging with AI Act, which would support research into AI’s effects on older adults, and his Senior Chatbot Protection Act, which would require clearer disclosures and safeguards from chatbot and voice-assistant providers. 

Kelly and Gillibrand also sought information from major AI companies about protections for older users. 

The hearing reflected bipartisan agreement that AI also has beneficial uses, including detecting suspicious transactions and supporting health and social services. But lawmakers said those opportunities cannot obscure the immediate danger posed by tools that allow scammers to manufacture trust, fear, and urgency with unprecedented realism.

For families, witnesses offered a simpler first line of defense: pause before acting, independently contact the person or institution supposedly seeking money, and establish a private family verification word for emergencies.