Anthropic Calls for Global Pause in AI Development, Warns of ‘Self-Improvement’ Risk

Anthropic had warned of a potential cybersecurity reckoning from frontier AI models when announcing its Claude Mythos Preview in April.

This photograph shows a smartphone displaying the logo of the US artificial intelligence safety and research company Anthropic, in Mulhouse, France, on April 21, 2026. (Sebastien Bozon/AFP via Getty Images

Artificial intelligence (AI) giant Anthropic on June 4 called for industry leaders worldwide to slow down the development pace of the technology, warning that the models are reaching a point where they could soon independently improve themselves, potentially taking control out of the hands of humanity.

Anthropic, maker of the popular Claude chatbot, described in a blog post how continuing to delegate the development of AI to the systems themselves risks creating a technology that is “capable of fully autonomously designing and developing its own successor.”