More than 20 industry leaders warned of the "intelligence explosion", and executives from Anthropic and OpenAI called for attention to AI self-improvement20 余名行业领袖示警“智能爆炸”，Anthropic、OpenAI 高管呼吁关注 AI 自我改进

ITHome News on September 29. According to a Bloomberg report on the 28th, executives from Anthropic, OpenAI, Meta and Microsoft jointly urged policymakers to pay close attention to the extent to which AI systems have improved themselves and take measures to ensure the security of AI technology. More than 20 AI industry leaders and researchers warn in a new paper that the rise of using AI to automate the development of next-generation models could lead to a sudden acceleration in AI technology progress in the future. The authors of the paper call this phenomenon the "intelligence explosion." Once this happens, the development speed of AI capabilities may exceed society's ability to guide and adapt to AI technology. Participating in writing the paper include Anthropic co-founder Jack Clark, OpenAI chief scientist Jakub Pachotsky, and AI pioneers Jeffrey Hinton, Joshua Bengio and others. R…