AI News Feed
Market watch
Batteries Cut Australia's Evening Gas Use by 67% as California Avoids Rolling Outages Since 2020AI Voice Scams Grow More Convincing as Familiar Voices Lose Power to Prove IdentityUSB-C vs. 3.5mm Headphone Jacks: Audio Quality Hinges on Where Conversion HappensTechRadar Article Links Leaked Meta VR Headset to Author’s Quest Pro HopesApple Opens iPhone 18 Pro, AirPods 5 and Apple Watch Pre-Orders as Retailers Discount New HardwareMicrosoft Warns of Sophisticated Passkey Phishing Campaign Targeting Cloud Accounts
Batteries Cut Australia's Evening Gas Use by 67% as California Avoids Rolling Outages Since 2020AI Voice Scams Grow More Convincing as Familiar Voices Lose Power to Prove IdentityUSB-C vs. 3.5mm Headphone Jacks: Audio Quality Hinges on Where Conversion HappensTechRadar Article Links Leaked Meta VR Headset to Author’s Quest Pro HopesApple Opens iPhone 18 Pro, AirPods 5 and Apple Watch Pre-Orders as Retailers Discount New HardwareMicrosoft Warns of Sophisticated Passkey Phishing Campaign Targeting Cloud Accounts
Cybersecurity

AI Voice Scams Grow More Convincing as Familiar Voices Lose Power to Prove Identity

AI voice cloning is making “relative in distress” scams harder to detect. A 2026 study found 36.1% of 4,100 US adults said they would comply or were unsure about such a scam, while Malwarebytes reported 85% of people struggle to tell scams from real contact.

AI News Feed · 2026-09-15 01:30

One X user described a call from someone who sounded like his wife, saying she had forgotten her wallet and needed a credit card to pay for gas. The user wrote that his wife was at home with him and drove a Tesla. “Same voice, bit... off, weird cadence, but 100% sounded just like her,” the user posted. “If she wasn't there, and it wasn't about gas I would have fallen for it.” The user said he kept the caller on the line to learn more and believed it may have been a voice filter, though TechRadar noted it was unclear whether the call used generative AI or a real-time voice filter.

When the story reached Reddit, some people were skeptical about whether it had happened, but others shared similar experiences. “This just happened to me last week,” one user said. “They called aunt and grandmother and said they were an attorney and that I was in jail and needed to be bailed out.” TechRadar reported that the “relative in distress” scam is an old trick, but AI is making the danger more real because a scammer may actually sound like a loved one.

The researchers also found that persuasiveness mattered more than how human the voice sounded, and that people who regularly used AI were not any better at identifying synthetic voices than people with no AI exposure. The study’s 36.1% figure measured self-reported willingness to comply, not the number of people actually fooled into handing over money.

Security advice includes setting up a family code word that everyone knows and a scammer is unlikely to guess or discover online. A simpler precaution is to hang up and call the person back on a number already known. If a son, daughter, partner or parent calls out of the blue asking urgently for money, the voice alone should not be treated as proof.

TechRadar cited a 2024 Ferrari deepfake case to show how personal verification can work. A Ferrari executive received messages followed by a phone call apparently impersonating CEO Benedetto Vigna. Suspicious of the call, the executive asked the supposed Vigna to identify a book the real CEO had recently recommended to him. The caller could not answer, and the attempted deception fell apart.

Mark Beare, Head of Consumer at Malwarebytes, said: “AI's deepest impact isn't on our devices; it's on us. When people can no longer trust what they see, hear, or who they're talking to, the damage reaches far beyond any single scam and into the building blocks of our society.” A Malwarebytes survey published in June found that 85% of people said it was hard to tell a scam apart from the real thing, up from 66% the previous year. 88% said it was becoming harder to tell which content online was genuinely human or real.

The problem is not limited to voice calls. AI-generated images and deepfake videos are increasingly challenging the assumption that seeing or hearing something is proof that it happened. TechRadar reported that a familiar voice used to be enough proof of identity, but in 2026 that is no longer enough. It advised agreeing on a code word with close contacts now and hanging up and calling back if an unexpected call asks for money.

Editor's Summary AI voice cloning is making “relative in distress” scams more convincing, with a 2026 study finding 36.1% of 4,100 US adults said they would comply or were unsure about such a scam. Security experts and a 2024 Ferrari deepfake case show that voice recognition alone is no longer reliable proof of identity. The recommended response is to set a family code word and verify unexpected money requests by calling the person back on a known number.