Researchers warn 'humans cannot reliably detect' audio deepfakes even when trained

France Nouvelles Nouvelles

Researchers warn 'humans cannot reliably detect' audio deepfakes even when trained
France Dernières Nouvelles,France Actualités
  • 📰 FoxNews
  • ⏱ Reading Time:
  • 40 sec. here
  • 2 min. at publisher
  • 📊 Quality Score:
  • News: 19%
  • Publisher: 87%

A U.K. study found humans can only accurately identify deepfake audio a quarter of the time, even when they are trained to differentiate the AI-generated audio. AI-generated audio that mimics humans can be so convincing that people can't tell the difference a quarter of the time – even when they're trained to identify faked voices, a new study claims.

humans can be so convincing that people can't tell the difference a quarter of the time – even when they're trained to identify faked voices, a new study claims.

"Previous literature has highlighted deepfakes as one of the biggest security threats arising from progress in artificial intelligence due to their potential for misuse," researchers wrote in their paper published this month. AI-generated audio that mimics humans can be so convincing that people can't tell the difference a quarter of the time.that generated 50 deepfake speech samples. The researchers used both English and Mandarin speech"t​​o understand if listeners used language-specific attributes to detect deepfakes."

"Our findings confirm that humans are unable to reliably detect deepfake speech, whether or not they have received training to help them spot artificial content," Kimberly Mai, an author of the study, said in a statement.

Nous avons résumé cette actualité afin que vous puissiez la lire rapidement. Si l'actualité vous intéresse, vous pouvez lire le texte intégral ici. Lire la suite:

FoxNews /  🏆 9. in US

France Dernières Nouvelles, France Actualités

Similar News:Vous pouvez également lire des articles d'actualité similaires à celui-ci que nous avons collectés auprès d'autres sources d'information.

How quickly can Utah’s Cam Rising get back to 100%? Here’s what researchers say about QBs and ACLsHow quickly can Utah’s Cam Rising get back to 100%? Here’s what researchers say about QBs and ACLsThe Utes' Cam Rising is trying to return from an ACL tear suffered just 8 months ago. How likely is he to succeed — and how good will he be when he comes back? A look at the academic research on ACL tears and football recovery.
Lire la suite »

Scientists Regrow Retinal Cells in The Lab Using NanotechnologyScientists Regrow Retinal Cells in The Lab Using NanotechnologyIn efforts to tackle the leading cause of blindness in developed countries, researchers have recruited nanotechnology to help regrow retinal cells.
Lire la suite »

Celine Dion still searching for effective treatment for 'horrid' neurological disorderCeline Dion still searching for effective treatment for 'horrid' neurological disorderDion\u0027s sister Claudette told a magazine that Celine is \u0027listening to the top researchers in the field\u0027 of stiff person syndrome, but has yet to find a treatment that works
Lire la suite »

A Va. research farm prepares to receive a key addition — a dead bodyA Va. research farm prepares to receive a key addition — a dead bodyIn a Prince William County forest, George Mason University researchers plan to study body decomposition to help advance criminal and other investigations.
Lire la suite »

AI can now steal your passwords with almost 100% accuracy | Digital TrendsAI can now steal your passwords with almost 100% accuracy | Digital TrendsResearchers have discovered a devious new way for AI to steal your passwords, and it's almost 100% accurate.
Lire la suite »

US uses lasers to achieve fusion energy gain for second timeUS uses lasers to achieve fusion energy gain for second timeUsing a 192-beam laser at Lawrence Livermore National Laboratory's National Ignition Facility, researchers heated and compressed hydrogen atoms, exceeding solar temperatures.
Lire la suite »



Render Time: 2025-03-01 18:33:21