Get all your news in one place.
100's of premium titles.
One app.
Start reading
TechRadar
TechRadar
Ellen Jennings-Trace

DeepSeek ‘incredibly vulnerable’ to attacks, research claims

A person using DeepSeek on their smartphone.

  • DeepSeek AI model performs poorly against its peers in safety testing
  • The R1 model exhibited a 100% attack success rate in Cisco testing
  • AI chatbots can be 'jailbroken' to perform malicious tasks
Sign up to read this article
Read news from 100's of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
One subscription that gives you access to news from hundreds of sites
Already a member? Sign in here
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.