[DeepMind Proposes International AI Model Review Organization, Musk and Altman Express Support] According to a report by Jinse Finance, on July 17, Google DeepMind CEO Demis Hassabis proposed the establishment of an international independent organization to conduct unified testing and review of advanced AI models before their release. He also plans to visit Washington next week to discuss related proposals with the U.S. government. Hassabis stated that the advanced cyberattack capabilities of Anthropic's Mythos model highlight the necessity of strengthening AI safety regulations. The proposal suggests that independent experts would be responsible for reviews, leading AI companies would provide funding, models would undergo testing for up to 30 days, and, if necessary, the industry would coordinate to slow down the development of advanced models. OpenAI CEO Sam Altman and Elon Musk have publicly expressed their support.