The Ban Artificial Superintelligence Act, explained
A bill in Congress would outlaw AI smarter than humans, pause the most advanced AI work and create a Department of Artificial Intelligence. Both versions have been sent to committee.
- What happened
- Sen. Bernie Sanders (I-Vt.) introduced the Ban Artificial Superintelligence Act of 2026 in the Senate on September 23, 2026 (S. 5493). Rep. Greg Casar (D-Texas) introduced it in the House the next day (H.R. 10538).
- Who it covers
- Anyone developing, deploying, funding or possessing very advanced AI in the US: in practice the frontier AI companies.
- What changes
- Nothing yet. A bill only becomes law if both chambers pass it and the President signs it. Both versions were referred to committee in September.
What happened
Sanders and Casar first announced the bill on September 3, 2026, as forthcoming legislation.4 On September 23, Sanders introduced it in the Senate as S. 5493, and it was referred to the Committee on Commerce, Science, and Transportation.1 On September 24, Casar introduced the House version, H.R. 10538, with nine original co-sponsors; it was referred to the Committee on Oversight and Government Reform.2
Announcing the introduction, Sanders said: “When you are racing towards a cliff, you don’t just ease up on the gas pedal. You hit the brakes.”3 Casar said the bill “bans the development of artificial superintelligence and pushes for international agreements so that no one, anywhere, builds AI too powerful for humans to control.”3
What the bill would do
A permanent ban: No one could develop, deploy, acquire, possess, fund, import or transfer “artificial superintelligence”.5 The bill defines it as an AI system that exceeds human cognitive performance across most domains or tasks, or that “has sufficient capabilities to plan and execute the destruction or disempowerment of humanity, including by overthrowing or undermining the Federal Government.”5 The ban also covers systems showing “precursor” abilities, such as automating AI research, hacking into systems, resisting shutdown, helping with nuclear, chemical or biological weapons, or deceiving human overseers.5
A pause on advanced AI: Systems trained with 10^25 or more operations of computing power would be paused until a new department is staffed and has set safety rules, including pre-deployment approval before any advanced system is released.51
A new Department of Artificial Intelligence: a cabinet-level department, led by a Secretary confirmed by the Senate and advised by a scientific advisory board, would monitor advanced systems, license AI companies through charters and order dangerous systems shut down.5
Penalties: individuals in policymaking roles, or rogue actors, who break the pause or ban could face up to 20 years in prison; other employees could be barred from the AI industry for 10 years; companies would lose their charter and surrender the systems involved, which the sponsors call a “corporate death penalty.”53 The bill also directs the US to seek international agreements and export controls to stop superintelligence being built elsewhere, and protects whistleblowers.5
What people are saying
The sponsors say AI companies cannot be trusted to slow down on their own. Sanders: “When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules.”3
The Information Technology and Innovation Foundation, a tech-policy think tank, opposed the proposal. Its president, Daniel Castro, said more powerful AI could speed up medical and scientific research, and warned that a unilateral ban could advantage China, since it would not stop advanced models being developed outside the United States.6
What it means for you
- 1This is a proposal, not a law. Nothing in the bill applies today. To become law it would have to pass both the Senate and the House and be signed by the President.
- 2It puts AI extinction risk on the record in Congress. The bill’s findings cite public warnings from AI leaders about losing control of AI. Our guide what p(doom) means explains those estimates.
- 3You can read the whole thing. The full text of S. 5493 and H.R. 10538 is public (see sources). Your senators and representative are the people who would vote on it.
What happens next
- 3 Sep 2026 · doneSanders and Casar announce the bill
- 23 Sep 2026 · doneS. 5493 introduced and referred to the Senate Commerce Committee
- 24 Sep 2026 · doneH.R. 10538 introduced and referred to the House Oversight Committee
Sources
- [1]S. 5493, Ban Artificial Superintelligence Act of 2026 (introduced in Senate) · U.S. Government Publishing Office · primary source · accessed 2026-10-15
- [2]H.R. 10538, Ban Artificial Superintelligence Act of 2026 (introduced in House) · U.S. Government Publishing Office · primary source · accessed 2026-10-15
- [3]Sanders, Casar Introduce Legislation to Create New Federal Agency to Ban Artificial Superintelligence, Pause Advanced AI Development (Sept. 23, 2026) · Office of Sen. Bernie Sanders · statement · accessed 2026-10-15
- [4]Sanders, Casar to Introduce Legislation to Ban Artificial Superintelligence and Temporarily Pause Advanced AI Development (Sept. 3, 2026) · Office of Sen. Bernie Sanders · statement · accessed 2026-10-15
- [5]Ban Artificial Superintelligence Act: section-by-section summary · Office of Sen. Bernie Sanders · primary source · accessed 2026-10-15
- [6]Sanders and Casar propose ban on artificial superintelligence (Sept. 4, 2026) · ForkLog · news report · accessed 2026-10-15