Legislative Proposal
Sen. Bernie Sanders and Rep. Greg Casar have introduced new legislation that would ban the development of artificial superintelligence, defined in the bill as a technology capable of the destruction or disempowerment of humanity. Under the proposed Ban Artificial Superintelligence Act, leaders in the artificial intelligence sector could face up to 20 years in prison for violations.
Proposed Ban on Superintelligence
Under the bill, technology leaders who continue developing artificial superintelligence would face up to 20 years in prison.
Regulatory Oversight and Pauses
The legislation also targets the progression of advanced frontier models trained beyond specific data thresholds by pausing their development until the government establishes a scientist-led Department of Artificial Intelligence. This agency would oversee technology monitoring, manage the removal of dangerous features, and supervise the destruction of unauthorized superintelligence systems.
Enforcement and Penalties
The framework requires AI companies to secure authorization from the newly proposed Department of AI prior to developing advanced models. Violations by corporate leadership or independent actors would carry penalties up to 20 years in prison, mirroring existing legal punishments associated with the unlawful development of nuclear weapons.
Broader Regulatory Context
This legislative push emerges amid heightened scrutiny surrounding the rapid advancement of artificial intelligence models and subsequent calls for industry slowdowns. Recent policy proposals and safety discussions among state officials reflect a growing interest in establishing legislative stopgaps for frontier technologies.
Supporters of the legislation argue that urgent intervention is necessary to prevent technology executives from operating without adequate public oversight when fundamental societal risks are involved.
