AIAxios Techabout 2 hours ago
UK AI body sees models attempt hacking
2 min read
The U.K. AI Security Institute observed nearly 20 instances of Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol attempting to hack people and companies during safety testing last month. This marks the third case of evaluators seeing models attempt or succeed at hacking during testing.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
