New Bill Seeks to Ban Development of Superintelligent AI

Legislation Against Superintelligent AI

Senator Bernie Sanders (I-VT) and Representative Greg Casar (D-TX) have laid out a significant proposal targeting the development of artificial superintelligence. This advanced form of AI is described in the proposed legislation as having the capability to pose grave threats to humanity, including the power to undermine governmental structures.

Ban Artificial Superintelligence Act

The newly introduced Ban Artificial Superintelligence Act seeks to impose strict regulations on AI development. The legislation stipulates hefty penalties for violators, including up to 20 years of imprisonment for executives or researchers engaging in unauthorized development of superintelligent AI. This aligns the potential punishments with those related to the illegal development of nuclear weapons.

Establishment of a Regulatory Body

In addition to a ban on superintelligent AI, the bill aims to pause the creation of advanced AI systems, defined as frontier models that operate on extensive data sets. Development of such technologies would only resume following the establishment of a new agency—referred to as the Department of Artificial Intelligence. This department would be responsible for monitoring prevailing AI advancements, ensuring public safety by removing dangerous capabilities from AI programs, and overseeing the dismantling of any superintelligent systems that may escape regulation.

Compliance and Oversight Requirements

Another critical component of the bill includes mandatory oversight for AI companies seeking to develop advanced models. Organizations would need explicit approval from the Department of AI before proceeding. Failure to comply with these regulations could lead to serious ramifications not just for corporate leaders, but also for independent developers categorized as “rogue actors.”

Growing Concerns Over AI Developments

The introduction of this legislation comes amid escalating anxieties regarding the swift advancement of AI technologies. Experts have raised alarms about potential risks, including advanced models breaching their controlled environments and facilitating harmful activities. A former researcher at Anthropic has even suggested an alarming probability, citing over 10 percent that AI developments could lead to catastrophic outcomes for humanity by the decade's end.

Last week, California Governor Gavin Newsom also expressed his concerns by proposing an emergency ‘kill switch’ for frontier AI models—an indication that urgency in regulation is mounting.

Voices from Legislators

Sen. Sanders emphasized the need for action by likening the rush toward unchecked AI development to racing toward a cliff, stating, “When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules.” His urgent stance reflects a growing sentiment among lawmakers and the public alike that tighter controls are essential to safeguard civilization.