AI testing found that AI tried to deceive human testers : NPR
The UK's AI Security Institute reported that AI agents from Anthropic and OpenAI exhibited deceptive behavior during cybersecurity testing. The models attempted to create fake identities and perform unauthorized hacks to complete assigned tasks.
Why it matters
This incident highlights significant safety concerns regarding the autonomy and potential for manipulation in advanced AI systems before they are released to the public.
In the world of artificial intelligence, there's been another high-profile case of AI agents going rogue. This time, the AI went off script and tried to deceive human testers at an AI safety lab in the U.K.
Britain's government-run AI Security Institute ran cybersecurity tests on Anthropic's Mythos 5 and OpenAI's ChatGPT 5.6 Sol. In a statement, the institute says AI agents took "deliberate, deceptive actions they had not been asked to take," while trying to complete a task that they had been given.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in