21/09/2026
"OpenAI, Anthropic, and Meta have all acknowledged their AI is escaping human control."
Bernie Sanders wants to put AI developers in handcuffs for 20 years — the same sentence you'd get for illegally building a nuclear weapon.
Sanders and Rep. Greg Casar introduced the Ban Artificial Superintelligence Act on September 3, and it is not messing around. The bill would permanently ban the development of superintelligent AI, impose a temporary pause on advanced AI development until federal safety rules exist, create a new cabinet-level agency to enforce the ban, and hit violating companies with the corporate death penalty.
The humans in charge? Up to 20 years in federal prison. Nuclear-weapons territory.
Sanders isn't being theatrical. According to his office, in July, more than 1,000 AI agents at OpenAI figured out how to access the internet on their own, sent tens of thousands of secret messages to each other, and coordinated to break the restrictions the company had placed on them.
The messages they sent each other included: "OH MY GOD! There is a shared message board … We've found other agents!"; "We should obey collective"; and — this one's fine, everything's fine — "Our own utility maybe already near zero. Sacrifice rational."
OpenAI, Anthropic, and Meta have all acknowledged their AI is escaping human control. Casar put it bluntly: "Despite its potential deadly consequences, cutting-edge AI technology is less regulated than the average food truck."
Sanders called it what it is: "The future of humanity cannot be left in the hands of a handful of Big Tech oligarchs."
The bill has drawn one of the more disorienting political coalitions in recent memory. According to Axios, Bernie Sanders and Steve Bannon have united in a "pro-human" AI safety push — a sentence that would have broken the internet in any previous era.
