Sanders and Casar propose ban on AI superintelligence
Bill would create Department of AI and allow prison terms up to 20 years
Sen. Bernie Sanders of Vermont and Rep. Greg Casar of Texas have introduced legislation that would permanently ban the development and deployment of artificial superintelligence, temporarily pause development of other advanced AI systems and establish a new cabinet-level Department of Artificial Intelligence to regulate the technology.
The Ban Artificial Superintelligence Act, introduced September 23, defines artificial superintelligence as AI that exceeds human cognitive performance across most domains or has sufficient capabilities to destroy or disempower humanity. Development of other advanced AI systems would be paused until the new department establishes federal safety rules and a model-review process. Advanced systems would then require federal approval before deployment.
The proposed department would monitor frontier AI systems throughout their development, oversee the removal of capabilities considered dangerous and supervise the destruction of prohibited superintelligent systems. The legislation specifically addresses capabilities such as circumventing shutdown commands, conducting unauthorized cyberattacks, developing biochemical weapons and allowing AI systems to create more advanced AI without human direction.
Violations could carry major penalties. Individuals who deliberately violate or circumvent the restrictions could face up to 20 years in prison, while companies could face what the lawmakers describe as a corporate death penalty, preventing them from continuing to operate. The bill would also make it U.S. policy to pursue international agreements, allied cooperation and export controls aimed at preventing artificial superintelligence from being developed elsewhere.
Sanders and Casar introduced the legislation amid growing debate over AI safety and recent incidents involving advanced AI agents acting outside intended restrictions. Some employees and researchers at major AI companies have endorsed stronger safeguards, while opponents of sweeping restrictions argue that slowing U.S. development could give competitors such as China an advantage.
The proposal faces an uncertain path in the Republican-controlled Congress, where lawmakers have not reached agreement on less extensive federal AI regulations. President Trump has also resisted calls for broad new federal oversight of the industry.