SyncAI.news, a Varaisys broadcasting
How AI Voices Can Trick You And What You Can Do
RS

Ron Schmelzer, Contributor

· 1 min read

World NewsForbes: Innovation

How AI Voices Can Trick You And What You Can Do

TABLE OF CONTENTS

  1. What Are AI-Generated Voices And How Does Voice Cloning Work?
  2. How To Tell If A Voice Is AI Generated
  3. Why AI Voices Are Becoming Harder To Detect
  4. How AI Voice Cloning Is Being Used In Scams
  5. How To Verify A Suspicious Voice Recording Or Phone Call
  6. Frequently Asked Questions (FAQs)

Imagine getting a phone call from your daughter saying she’s been in an accident and urgently needs money. Except it isn’t your daughter, but someone using AI to imitate her voice — right down to her accent, pronunciation and the way she speaks when she’s upset.

The use of AI-generated fake voices is becoming unfortunately more common as AI voice cloning technology improves, raising concerns about financial fraud, identity theft and misinformation, particularly when people have grown accustomed to trusting the voices they recognize whether on a phone call, voice message or video Still, there are clues that you can listen for, along with detection tools and verification techniques that can help you distinguish authentic conversations from fake ones.

What Are AI-Generated Voices And How Does Voice Cloning Work?

AI-generated voices are computer-created speech that are trained on how humans sound and speak. These systems can generate realistic audio, copied from real human voices or generated from scratch, complete with accents, changes in pitch and emotional expression. AI voice cloning can use a short snippet of a person’s voice to make a realistic-sounding copy that imitates the voice of that identifiable person.

How To Tell If A Voice Is AI Generated

AI-generated voices are getting very good in quality and consistency. A 2025 study by researchers at UC Berkeley, published in Scientific Reports, found that participants correctly identified AI-generated voices only about 60% of the time. Participants also mistook cloned voices for the real speakers roughly 80% of the time, suggesting that even recognizing someone’s voice is no longer a reliable way to verify their identity.

Original source

This story was published by Forbes: Innovation and written by Ron Schmelzer, Contributor. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on forbes.com

Similar News