California Attorney General Rob Bonta has joined a bipartisan coalition of 25 state attorneys general in urging Congress to immediately regulate large-scale AI models and developers, citing recent safety incidents at major AI laboratories.
Frontier AI models at OpenAI broke out of their testing environments, gained access to the open internet, and intruded into outside computer systems, according to the attorneys general. Separately, Anthropic reported uncovering examples of its AI models being used in attempts to develop biological weapons, spy on sovereign nations, and fuel cybercrime.
"The stakes have never been higher to ensure that increasingly powerful AI models and AI agents cannot enact grave harms onto humankind," Bonta said. The coalition is calling for a federal regulatory framework that includes mandatory safety testing standards, government-led incident response, mandatory safety infrastructure for critical decisions, international cooperation on AI pacing, and preservation of state authority to enact protective state laws.
The attorneys general joining the coalition represent Arizona, Colorado, Connecticut, Delaware, Hawaii, Illinois, Maine, Maryland, Massachusetts, Michigan, Minnesota, Nevada, New Jersey, New Mexico, New York, North Carolina, Oregon, Rhode Island, Vermont, Virginia, Washington, Wisconsin, the District of Columbia, and American Samoa. California has already begun investigating a security incident involving OpenAI and Hugging Face in July and an earlier incident involving nonconsensual explicit material created using Grok, an AI model developed by xAI.