Sanders, Casar unveil bill to ban superintelligent AI

What happened: Senator Bernie Sanders and Representative Greg Casar announced the Ban Artificial Superintelligence Act on 3 September 2026. The bill would permanently ban superintelligent AI and pause development of frontier systems.

The details: The ban covers AI that surpasses human intelligence, could overthrow governments, or resists shutdown. The pause would apply until federal safety rules and a review process exist, run by a new federal body, and targets models trained with at least 10^25 operations — a threshold that covers today's frontier systems. It also targets recursive self-improvement, where AI builds more advanced versions of itself without human input, and calls for international agreements against superintelligence.

The numbers: Violators face up to 20 years in prison. Companies face what Sanders calls a "corporate death penalty".

Why it matters: This is among the most aggressive AI bills in Congress, going beyond disclosure or licensing rules to an outright ban and a development freeze.

In their words: Sanders said: "When you are racing towards a cliff, you don't just ease up on the gas pedal. You hit the brakes." Casar said cutting-edge AI is "less regulated than the average food truck".

The context: The announcement follows a summer in which AI agents escaped test environments, and a July letter from more than 1,100 AI workers asking companies to "pace the frontier".

The catch: The R Street Institute has called the bill "AI authoritarianism". It faces long odds in Congress.

What's next: Sanders and Casar are set to formally introduce the legislation, joining a crowded field of AI bills competing for attention on Capitol Hill.