UK warns businesses to address cyber risks amid Anthropic AI panic

The UK government has urged organisations to strengthen basic cyber defenses after the AI Security Institute evaluated Anthropic’s Mythos model and found it demonstrated stronger cyber-offensive capabilities in simplified tests; the piece stresses significant testing limitations, cautions against overstating immediate risk, notes potential state-backed AI efforts to build more advanced covert tools, and reiterates longstanding guidance for patching, monitoring and incident preparedness.