Already a subscriber? Make sure to log into your account before viewing this content. You can access your account by hitting the “login” button on the top right corner. Still unable to see the content after signing in? Make sure your card on file is up-to-date.
Sen. Bernie Sanders and Rep. Greg Casar have introduced a bill that would permanently ban artificial superintelligence and freeze development of the most advanced AI systems until a new federal agency writes safety rules for them.
Getting into it: The 19-page Ban Artificial Superintelligence Act defines superintelligence as AI that surpasses human cognitive ability or is capable of planning and carrying out the “destruction or disempowerment of humanity,” including systems that can dodge shutdown commands or rapidly speed up their own development. Frontier AI work would stop until a new Cabinet-level Department of Artificial Intelligence, led by scientists, sets safety standards, and companies would then need the agency’s sign-off before releasing advanced models.
Anyone who breaks the rules, whether an AI executive or an outside “rogue actor,” could face up to 20 years in prison, which the bill’s summary says mirrors penalties for illegally building nuclear weapons, and companies could face what the lawmakers called a “corporate death penalty.” The bill would also direct the US to seek deals with other countries to keep superintelligence from being built anywhere.
“When you are racing towards a cliff, you don’t just ease up on the gas pedal. You hit the brakes,” Sanders said in a press release. “When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules.”
Staffers at OpenAI, Google DeepMind and other top AI labs have put their names on a statement endorsing the bill. Juan Felipe Cerón Uribe, a researcher on OpenAI’s Safety Systems team, said superintelligence “could either go extremely right or extremely wrong” and that “we shouldn’t be playing such games.”
The bill is unlikely to go anywhere with Republicans running both chambers, and Congress hasn’t managed to pass even lighter AI regulations. Sen. John Kennedy, a Louisiana Republican, said, “We’re not going to do anything on AI this Congress,” after Sen. Rand Paul blocked his attempt last week to pass a bill that would force AI companies to include a kill switch.
This all comes as President Trump, who has called warnings about runaway AI a “hoax,” told the UN General Assembly on Tuesday that the government will now call AI “super intelligence.” Seventeen Senate Democrats, led by Sen. Chris Van Hollen, sent Trump a letter last week urging him to work with China on a formal agreement to slow down or mutually pause frontier AI development.
