ai
Wired Ai
2026-08-27
Anthropic outlines safety framework for AI agents in the physical world
AI research company Anthropic emphasizes that the benefits of automating scientific and industrial tasks must be weighed against emerging security risks.