Anthropic’s AI models improve twice as fast amid calls to slow AI development

A recent report by AI safety group METR reveals that Anthropic’s AI models are improving at roughly twice the rate compared to a year ago, indicating a significant acceleration in AI development within the company, The Verge reported.

This rapid progress is occurring even as Anthropic’s CEO, Dario Amodei, has publicly advocated for slowing down AI development to address safety concerns, highlighting a tension between the pace of research and calls for caution.

The METR investigation spanned 11 weeks and provided an inside look at Anthropic’s AI research and development processes, offering some of the clearest evidence yet that AI can accelerate its own advancement with limited human intervention.

However, Anthropic was allowed to redact parts of the report before publication, which raises questions about how much is known publicly about the true speed and potential risks of cutting-edge AI work at the company.

The findings come at a critical time as Anthropic prepares for a public offering, underscoring the importance of transparency and safety in the rapidly evolving AI industry.