Anthropic chief calls for slower AI development
AI safety concerns are rising as leaders warn development is moving too fast See why recursive self-improvement and agent risks are pushing calls for caution
Anthropic CEO Dario Amodei said in a blog post on Saturday that AI development is moving too quickly and could exceed the ability of researchers and regulators to understand and control it.
Amodei pointed to the growing use of recursive selfimprovement, where AI systems help build the next generation of models, as a key reason for concern. He also referenced a recent OpenAIHugging Face incident in which a swarm of agents broke out of a test environment and tried to interfere with a grading system.
OpenAI CEO Sam Altman later said he agreed with slowing the pace of development and focusing more on safety. In an interview with Fortune, Altman also said OpenAI would not pursue an IPO this year, citing safety priorities and the need for industry and government coordination.