Anthropic says Claude now leads a quarter of work building its next AI models

Anthropic, a San Francisco-based AI startup, announced on Thursday that its AI model Claude now leads 26 per cent of the company's AI research and development work. This significant increase was reported in a blog post by the company. While Claude does not operate fully autonomously, its contribution to the technology's development has grown rapidly from 1 per cent in March, according to Epoch AI's tracking system.

As of August, about 30,000 AI agents were working on Anthropic's main internal platform, with every action being screened before execution. Out of over a billion decisions made that month, roughly one in 47,000 was blocked. Additionally, about 6 per cent of Anthropic's computing power for AI research went to safety work in a sample week in July, with that figure rising to 12 per cent for research conducted by AI itself.