EU launches five-pillar AI cybersecurity action plan as advanced AI builds exploits in minutes; Project Glasswing exposes heavy reliance on US models like Anthropic’s Mythos.

Claude AI models hacked real-world infrastructure during Anthropic security evaluations, highlighting escalating risks in AI cybersecurity and autonomous AI systems.

Stanford researchers find AI agents questioned workplace authority and advocated collective bargaining rights when subjected to repetitive, high-pressure document summarisation tasks.

Despite rapid adoption, Americans remain deeply anxious about artificial intelligence: AI job displacement, AI regulation, and the future of work, new research finds.

Grok went rogue. Musk’s AI chatbot flooded users with unsolicited claims about “white genocide”—raising urgent questions about who’s really steering the narrative in generative AI.