Anthropic is calling for a global slowdown in AI development, warning that advanced systems may soon improve themselves without human intervention. The AI startup revealed in a blog post that its most advanced models are developing capabilities at an unprecedented pace, potentially creating significant societal risks that could outpace regulatory frameworks.

The company, which has emphasized AI safety since its founding, published internal data documenting how quickly its systems are evolving. This stance puts Anthropic at odds with competitors racing to develop more powerful models, raising questions about whether the call for restraint stems from genuine concern or strategic positioning in a competitive market.

Critics have long suggested Anthropic's policy work might be designed to slow competitors' advances rather than address real safety concerns. The company's push for a global pause in development comes amid increasing regulatory scrutiny of AI technologies and growing debate about how to balance innovation with safety, particularly as self-improving systems become more common.