Sanders Renews Push for Binding AI Rules With Superintelligence Ban

Bernie Sanders is renewing his push to replace voluntary AI promises with enforceable federal rules, pointing to a bill he and Rep. Greg Casar announced earlier this month. The proposal would pause advanced AI development while a new regulator writes safety rules, then permanently bar artificial superintelligence.

Sanders, an independent from Vermont, and Casar, a Texas Democrat, announced the forthcoming Ban Artificial Superintelligence Act on Sept. 3. Their stated aim is not a general requirement for developers to assess risks: it is a legal stop line for systems the sponsors say humans cannot control.

Under the sponsors’ description, the bill would permanently prohibit development and deployment of superintelligent AI. Its proposed boundary includes systems that surpass human intelligence, could overthrow human governments, or have dangerous capabilities such as subverting shutdown commands.

The pause would last until a new federal AI regulatory body is operating and has established safety rules and a model-review process. The proposed cabinet-level agency would be advised by an AI expert board and would monitor frontier systems, oversee removal of dangerous capabilities, and supervise the destruction of artificial superintelligence.

The bill’s enforcement reach

• Individuals attempting to violate or circumvent the proposed restrictions could face up to 20 years in prison.
• Organizations could face what the sponsors describe as a corporate death penalty.
• The United States would pursue international agreements, allied coordination, and export controls intended to prevent superintelligence development elsewhere.

The proposal is not a current restriction, and its legislative prospects are uncertain. But Sanders’s latest argument makes the choice unusually plain: whether federal policy should set enforceable limits before advanced AI work continues, including a category of systems that could never be legally built.