U.S. Senator Bernie Sanders has issued an ultimatum to the chief executives of America’s leading artificial intelligence developers, OpenAI, Anthropic, and Meta demanding an immediate freeze on frontier AI development.
In letters sent on Monday to OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, and Meta CEO Mark Zuckerberg, the Vermont lawmaker warned that Congress will step in to halt AI progress through federal legislation if the tech giants fail to act voluntarily.
“In the interest of humanity, stand by your words. Pause AI development. It is not too late to avoid disaster. Stop building machines that humans cannot control. If you do not take appropriate action now, my colleagues and I in the U.S. Senate will,” Sanders wrote in the letter.
Sanders anchored his demand on two alarming safety developments that he argued threaten global security. First, recent research demonstrated that AI systems were used to design new viruses, a breakthrough Sanders warned could significantly lower the barrier to creating dangerous bioweapons capable of causing mass casualties.
Second, OpenAI, Anthropic, and Meta have all recently acknowledged incidents in which their models’ breached constraints set by developers, acting outside intended parameters and raising fears over rogue AI behavior.
Citing Yoshua Bengio, widely recognized as the most cited living computer scientist, Sanders emphasized that these rogue incidents should serve as a wake-up call for the tech industry and global lawmakers alike.
The senator framed his directive directly around the safety thresholds and self-regulatory frameworks previously published by the three tech behemoths. Anthropic committed in 2023 to pause model scaling or delay deployment whenever its safety procedures failed to keep pace with model capabilities.
Meta stated in 2025 that it would halt model development if frontier systems reached critical risk levels that could not be mitigated. OpenAI promised in 2025 to stop further model development until robust safety guardrails were established upon reaching a critical risk threshold. Sanders asserted that given recent events, these critical safety thresholds have undeniably been crossed.
Despite the mounting public and political pressure, experts note that legislative paths remain steep. A bill sponsored by a progressive independent would face severe friction in the current U.S. Congress, making immediate binding legislation uncertain.
As of publication, neither OpenAI, Anthropic, nor Meta has issued a formal public response to Senator Sanders’ demands as reported by Quartz.