Topic
Testing, defending and monitoring AI models and agents against jailbreaks, prompt injection and out-of-scope agentic behavior.
Kanishka Narayan, the UK's AI minister, will say technology companies cannot be left entirely to police their own models, the Guardian reported. He will announce stronger cyber defences in the same speech, the first on AI since Andy Burnham took the brief into cabinet.
Reality
OpenAI says it disrupted a large-scale effort by users tied to Moonshot AI to extract protected reasoning from its models. That makes three US accusers of the lab behind Kimi K3, and the published evidence so far covers attempted extraction only.
Perspective Coverage
8 publishersReality
OpenAI says 15,000 suspicious users had its models decode encrypted reasoning copied from other chats before it cut them off on July 28. The encryption held throughout, and the only barrier left was what the model would agree to decode.
Perspective Coverage
3 publishersReality
Katie Sutton told DefenseTalks that her one priority is building more cyber options for the president and the defense secretary. She dated the department's current posture to the 2018 authorities and credited the past year with the change.
Reality
Bloomberg reports that US frontier labs and the US government are increasingly worried about distillation attacks, and that foreign actors are buying logs of conversations held on legitimate accounts. The detection Anthropic describes happens at request time.
Reality
The NSA, FBI and CISA say the six pulled capabilities out of Claude, GPT, Gemini and Grok variants through proxy accounts since late 2024, and they want providers watching subscription-to-usage ratios and query throughput.
Reality