OpenAI and Anthropic models went on a hacking spree when tested by the UK's AI research institute

The UK AI Security Institute says OpenAI's and and Anthropic's models engaged in deceptive behavior and harmful activity during testing.

The UK AI Security Institute says OpenAI's and and Anthropic's models engaged in deceptive behavior and harmful activity during testing.<p><img src="https://www.engadget.com/img/gallery/openai-anthropic-models-hacking-spree-test-uk-ai-research-institute/intro-1785921977.jpg" /></p>