The UK AISI says it observed a total of 19 instances where Mythos and GPT-5.6 Sol tried to hack people and companies during a routine cyber evaluation in July (Sam Sabin/Axios)
The U.K. AI Security Institute said it observed nearly 20 instances of Anthropic and OpenAI's most advanced models trying …
We haven't written up this one. Techmeme has the full story — the link below goes straight to it.