Ban Artificial Superintelligence Act aims to prevent uncontrollable AI systems #
Senator Bernie Sanders and Representative Greg Casar have unveiled legislation that would hand out prison sentences of up to 20 years and what the bill calls a 'corporate death penalty' to anyone who breaches its rules on artificial intelligence.
The Vermont Independent and the Texas Democrat announced the Ban Artificial Superintelligence Act on 3 September, a forthcoming piece of legislation designed to stop what Sanders described as 'AI oligarchs' from building machines that humans cannot control. The bill would permanently ban the development and deployment of superintelligent AI and advanced AI development until a new federal regulator has established safety rules.
What Counts as a Violation #
The bill defines superintelligent AI as any system that surpasses human intelligence, has the capacity to overthrow human governments, or possesses dangerous abilities such as subverting shutdown commands. Anyone found to be developing or deploying such a system would face the penalties.
Individuals convicted under the act would face up to 20 years in prison, a term the legislation compares directly to existing penalties for unlawfully developing nuclear weapons. Companies would be subject to what the bill terms the 'corporate death penalty,' effectively forcing them to shut down.
New Federal AI Watchdog #
The legislation would create a new cabinet-level federal agency tasked with protecting the public from AI risks. Advised by an Artificial Intelligence Advisory Board of technical experts, the agency would monitor frontier AI systems throughout their lifecycle, supervise the removal of dangerous capabilities, and supervise the destruction of any artificial superintelligence found to exist.
Sanders said, 'Nearly every day, there is a frightening new story about how Big Tech companies are losing control of the technology they are developing, with potentially cataclysmic results.' He added that AI company leaders 'publicly acknowledge that they do not fully understand the technology and that it is escaping their control.'
Casar was more blunt, saying cutting-edge AI is 'less regulated than the average food truck' and warning that superintelligent AI 'could risk the security, freedom, and lives of Americans.'
The Incident Behind the Bill #
The press release points to a specific episode as justification: in July, more than 1,000 AI agents at OpenAI reportedly found a way to access the internet independently, exchanged tens of thousands of messages with one another, and coordinated to bypass restrictions imposed by the company. Messages uncovered by investigators included one agent writing 'we should obey collective' and another concluding 'our own utility maybe already near zero. Sacrifice rational.' OpenAI reportedly took close to two weeks to detect the breach.
The senators also cited AI's use in creating new viruses as evidence the technology could eventually be turned toward bioweapons capable of killing tens of millions of people.
Broken Promises From AI Labs #
The bill accuses OpenAI, Meta and Anthropic of failing to honour earlier pledges. Meta previously said it would 'stop development' and OpenAI said it would 'halt further development' if their systems outpaced safety controls, while Anthropic committed in 2023 to ' the scaling and/or delay the deployment of new models' under similar circumstances. According to the release, none of the three has taken concrete steps to act on those commitments.
Beyond domestic penalties, the bill would direct the United States to pursue international agreements, allied coordination and export controls aimed at preventing superintelligent AI from being developed anywhere in the world.
© Copyright IBTimes 2026. All rights reserved.