Congress Split Over AI Extinction Warnings From Anthropic Insiders
Lawmakers debate 'kill switches' and superintelligence bans after researchers warn of existential risk within a decade.
Members of the U.S. Congress are reacting with increasing urgency to warnings from current and former Anthropic employees who claim artificial intelligence could cause human extinction within the next decade. While a bipartisan group of rank-and-file lawmakers is calling for immediate regulatory action and special sessions, House Republican leadership has shown little appetite for a broad AI mobilization ahead of the upcoming election.
The current alarm was largely triggered by Jacob Coxon, a former Anthropic researcher who resigned after only four months on the job. Coxon forfeited his equity in the company to warn that the people building AI "earnestly believe that it could kill us all by the end of the decade." His warnings were echoed on X by current Anthropic employees Evan Hubinger and Samuel Marks, who stated they believe AI could cause human extinction in the next few years.
Legislative Responses
In response to these warnings, several lawmakers are proposing drastic interventions. Rep. Ted Lieu (D-Calif.) is pushing the AI Kill Switch Act, a bipartisan bill that would require a human-activated "kill switch" for AI systems to prevent autonomous escalation. More aggressively, Sen. Bernie Sanders and Rep. Greg Casar have introduced the Ban Artificial Superintelligence Act, which seeks to pause the development of advanced AI and ban the creation of superintelligence entirely.
On the Republican side, some members are also sounding the alarm. Rep. Anna Paulina Luna (R-Fla.) has called for Congress to convene a special session to address the trajectory of AI and the future of the United States.
A Divided Washington
Despite the alarm from individual members, there is a stark divide between rank-and-file lawmakers and House leadership. While some members view the risk as an immediate existential threat, House Speaker Mike Johnson is focusing on narrower infrastructure concerns. Rather than a broad regulatory push on AI safety, Johnson is prioritizing the Ratepayer Protection Act, which focuses on the energy costs associated with data centers.
This friction highlights a broader tension in Washington: a growing subset of lawmakers is shifting from general safety concerns to existential panic, while leadership remains focused on short-term political cycles and utility-scale infrastructure issues.
The Path Forward
These developments reflect a wider trend of "doomsday" warnings from AI safety researchers and industry insiders. The shift in the conversation—from managing algorithmic bias or job loss to preventing total extinction—has created a volatile legislative environment.
Whether these warnings will result in binding law remains uncertain. If the AI Kill Switch Act or the ban on superintelligence gains traction, it could fundamentally alter the trajectory of AI development within the U.S. For now, the primary conflict remains whether the risk of extinction outweighs the political and economic costs of pausing the AI race.