Anthropic's AI-driven development process doubles improvement pace, raising safety concerns

Anthropic is leveraging AI systems to automate parts of its own AI development process, and the results are significant. According to a report from METR, an AI safety research organization that spent 11 weeks investigating the company, Anthropic's models are now improving at roughly twice the speed they were a year ago, with no clear evidence of deceleration.

The findings highlight how rapidly AI development itself is being automated. The ability to use AI to improve AI introduces both capability and safety questions for the field, particularly around whether human oversight can keep pace with accelerating improvements.