Sanders and Casar introduce a bill to ban superintelligence outright

Senator Bernie Sanders and Representative Greg Casar, who chairs the Congressional Progressive Caucus, formally introduced the Ban Artificial Superintelligence Act on Wednesday. Sanders announced his intention to file it on 3 September; this week is the introduction itself.

The permanent ban covers developing or deploying an “artificial superintelligence”, defined in the sponsors' summary as a system that exhibits — or can easily be modified to exhibit — capabilities matching or exceeding human cognitive performance across most domains, or that can plan and execute the disempowerment of humanity or the overthrow of the federal government. Alongside it sits a temporary pause on advanced AI development generally.

What lifts the pause is the interesting part. It does not expire on a date. It lifts only once a new cabinet-level Department of Artificial Intelligence exists, is operational, and has issued safety rules and a model-review process. That department would monitor frontier systems across their lifecycle, supervise the removal of dangerous capabilities such as shutdown resistance, and oversee the destruction of any superintelligence. Deployment would additionally require its pre-approval. The State Department would be directed to pursue international agreements and export controls against superintelligence development abroad. Individual penalties run to 20 years.

Its prospects are poor. Fortune describes “long odds in the Republican-controlled Congress, where lawmakers have struggled to coalesce around even less sweeping AI regulations”, and the Alliance for Secure AI credited the sponsors for taking superintelligence seriously while urging a more measured approach rather than endorsing the bill. No bill number has appeared in either sponsor's materials, and no co-sponsors beyond the two principals are listed.

• Confirmed Introduced 23 September by Sen. Bernie Sanders (I-VT) with Rep. Greg Casar (D-TX); permanently bans developing or deploying artificial superintelligence and temporarily pauses advanced AI development. Office of Rep. Greg Casar
• Confirmed The pause lifts only once a cabinet-level Department of Artificial Intelligence is operational and has issued safety rules and a model-review process; maximum individual penalty is 20 years. Sanders bill summary (PDF)
• Reported Faces “long odds in the Republican-controlled Congress”; no bill number has been published by either sponsor, and no co-sponsors beyond Sanders and Casar are listed. Fortune