Connect with us

Hi, what are you looking for?

Tech & Science

AI voice cloning scams: Are they on the rise?

Search interest in “AI voice cloning scams” has surged 130% in the past month, signalling rapidly growing public concern.

Anyone with a smartphone and specialized software can create the harmful deepfake images
Anyone with a smartphone and specialized software can create the harmful deepfake images - Copyright AFP Mark RALSTON
Anyone with a smartphone and specialized software can create the harmful deepfake images - Copyright AFP Mark RALSTON

Image the situation: Your phone rings. It sounds like someone you trust. You start to reveal personal data. That’s exactly why voice cloning scams are working.

AI scam content has already generated 15.5 billion views across 4,500 videos, with creators focusing heavily on deepfake voice impersonation, family scams, and urgent “call” scenarios.

Unlike phishing emails or fake links, these scams rely on familiarity. A known voice lowers defences instantly and creates urgency before logic kicks in.

Search interest in “AI voice cloning scams” has surged 130% in the past month, signalling rapidly growing public concern. At the same time, short-form video platforms are amplifying awareness and fear around these scams.

This rise in activity (both scams and consumer interest) is according to new data from Virlo.ai.

The scale is accelerating, in that:

  • Around 1 in 4 adults globally report encountering AI voice scams
  • Deepfake-related fraud losses have surpassed $1.5 billion globally
  • Scammers now need just seconds of audio to clone a voice, often pulled from social media

But beyond such numbers, what is changing is accessibility. This is no longer a niche cybercrime. This nefarious activity is becoming mainstream.

According to Olga Scryaba, Head of Product at the firm isFake.ai, AI voices do not fail because they sound fake. They fail because they sound too clean.

“People expect robotic errors, but today’s AI fails in subtler ways. It’s the absence of human imperfection that gives it away”, she tells Digital Journal.

Five Signs a Voice Might Be AI

  1. Speech that’s too smooth, like no hesitation, no self-correction
  1. Emotion that doesn’t fully match the situation
  1. Overly consistent rhythm or pacing
  1. Slight glitches on names or uncommon words
  1. A subtle “off” feeling you can’t quite explain

Why Traditional Safeguards Are Failing

Voice used to be a trusted layer of verification. Now it is becoming a vulnerability.

Financial institutions are already rethinking voice authentication systems, while businesses face rising cases of executive impersonation scams used to authorize payments.

“The biggest risk is overconfidence,” Scryaba adds. “People assume they’ll recognize a fake voice. But these scams work because they don’t feel fake at the moment.”

How to Protect Yourself

According to Scryaba , simple habits can stop most attacks:

  • Set a family safe word for emergencies
  • Always verify through a second channel
  • Be mindful of how much voice content you share publicly

“Most people don’t fall for these scams because they’re careless,” explains Scryaba. “They fall for them because the situation feels urgent and emotionally real. That’s exactly what the scammer is counting on.”

Avatar photo
Written By

Dr. Tim Sandle is Digital Journal's Editor-at-Large for science news. Tim specializes in science, technology, environmental, business, and health journalism. He is additionally a practising microbiologist; and an author. He is also interested in history, politics and current affairs.

You may also like:

Business

America is made of ideas. It’s the lack of ideas that’s killing it.

Social Media

Elon Musk has been summoned for a voluntary interview in Paris as part of a French probe into his social media platform X.

Tech & Science

Pharmaceutical organizations must move from trust in AI based on early success to evidence‑based trust.

Life

Prince Edward Island is the worst area for crashes in Canada, with 14.1 car crash fatalities per 100,000 licensed drivers.