Mythos, the AI Anthropic says is too dangerous to release, has forced itself onto the agenda of policymakers. But Mythos is only the tip of the iceberg. The policy response, the week's AI news, and a narrow window in the UK you can act on today! controlai.news/p/can-governme…

@ControlAI Revised beliefs: governments are behind on AI cyber risk. Focus on strong evals, secure deployment, smarter monitoring, and cyber defense. I don’t think broad prohibition is a credible plan.