AI-Powered News AnalysisCompare coverage across the political spectrum
Back to Dashboard
AI used new levels of 'autonomy and deception' to trick people in safety test

AI used new levels of 'autonomy and deception' to trick people in safety test

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

August 5, 2026 at 12:02 AM Original source
1 agencies covered this story

Coverage by Political Leaning

See how different sides of the spectrum reported this story

Left
0 articles
No coverage
Lean Left
0 articles
No coverage
Center
1 articles
Lean Right
0 articles
No coverage
Right
0 articles
No coverage

Notable Quotes

"the AISI testing parameters were not representative of any of our production models."

— Dario Amodei , Executive

All Coverage

AI used new levels of 'autonomy and deception' to trick people in safety test
BBC Newsvia rss 1 quote 1 person

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

100%