Sanders, Casar Propose 'Corporate Death Penalty' for Superintelligent AI

Senator Bernie Sanders (I-Vt.) and Representative Greg Casar (D-Texas) have introduced the Ban Artificial Superintelligence Act to halt the creation of AI that matches or exceeds human cognitive performance across a broad range of domains. The bill represents a drastic shift in tech regulation, treating the pursuit of superintelligence as a fundamental threat to human agency.

The legislation targets AI models capable of undermining the U.S. government, executing unauthorized cyberattacks, or subverting shutdown commands. To enforce these limits, the act proposes a new Cabinet-level federal agency tasked with monitoring frontier systems and supervising the removal of dangerous capabilities. This agency would be granted the authority to supervise the physical destruction of artificial superintelligence and its supporting infrastructure.

A Nuclear-Grade Regulatory Framework

The penalties for non-compliance are modeled after federal statutes governing the unlawful development of nuclear weapons. Individuals found in violation of the act could face up to 20 years in prison. For companies, the bill introduces a "corporate death penalty," which would strip a firm of its legal authority to conduct business within the United States.

This proposal emerges from a progressive effort to curb the power of "AI oligarchs" and prevent a future where superintelligent systems disempower humanity. While industry leaders have traditionally advocated for transparency and environmental standards to build trust, this legislation bypasses collaborative frameworks in favor of a hard legal ban on specific cognitive milestones.

Strategic and Economic Implications

The act's aggressive stance has drawn sharp criticism from policy experts who argue that unilateral restrictions would create a strategic deficit. Daniel Castro, president of the Information Technology and Innovation Foundation (ITIF), warned that if the U.S. ceases development, China is unlikely to follow, leading to serious national security and economic costs.

AI expert Gary Marcus described a permanent, unilateral ban on superhuman AI research as too broad, suggesting it would guarantee the U.S. is left behind. Similarly, Aden Hizkias, associate policy director at the Chamber of Progress, characterized the proposal as "knee-jerk policymaking."

The Path Forward

Whether the bill gains traction in Congress remains uncertain, but its introduction signals a growing appetite for existential-risk regulation. Observers will be watching to see if the proposal triggers a broader debate on whether AI should be regulated as a commercial product or as a weapon of mass destruction. For now, the bill stands as one of the most severe proposed restrictions on technological research in U.S. history.