Latest Research: The safety supervision of top AI companies such as OpenAI does not meet global standards.

The updated version of the "AI Security Index" released by the non-profit Institute for the Future of Life on Wednesday pointed out that the safety governance level of major artificial intelligence companies such as Anthropic, OpenAI, xAI, Meta, etc., has not yet reached the emerging global standards. The institute stated that safety assessments conducted by independent expert groups found that despite these companies competing to develop "superintelligence," none has developed sufficiently comprehensive strategies and plans to control such advanced AI systems.