Prominent American politician Senator Bernie Sanders has introduced legislation to permanently ban artificial superintelligence, temporarily pause the development of advanced AI models, and establish a new cabinet-level Department of Artificial Intelligence to regulate the technology.
The legislation defines artificial superintelligence as any AI system that exceeds human cognitive performance across most domains or possesses capabilities sufficient to destroy or disempower humanity, including by overthrowing or undermining the federal government.
Part of the legislation reads,
Register for the next Tekedia Mini-MBA.
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
“When you are racing towards a cliff, you don’t just ease up on the gas pedal. You hit the brakes. When the future of humanity is at stake, we cannot let a handful of Big Tech CEOs write their own rules. Scientists have been clear: AI is an existential threat to humanity. It must be treated as such.
“This legislation will permanently ban the development of artificial superintelligence and would put in place an immediate pause on advanced AI development until we have clear safety rules in place developed by the most knowledgeable scientists in our country. Congress must act now before it is too late.”
Under the bill, no person or entity would be permitted to develop, deploy, acquire, possess, fund, import, or transfer such systems.
It also imposes an immediate pause on developing advanced AI systems until the proposed Department of Artificial Intelligence is fully operational and has established rigorous safety rules and model review processes.
The new department would serve as a cabinet-level agency tasked with safeguarding the public from AI-related risks.
Its responsibilities would include monitoring frontier AI systems throughout their lifecycle for dangerous capabilities, supervising the removal of hazardous features such as the ability to subvert shutdown commands, and overseeing the destruction of any systems that qualify as artificial superintelligence.
An advisory board of leading scientists and technical experts would provide independent guidance to the department. Sanders framed the measure as an urgent response to what he described as an existential threat.
Violations of the ban or the development pause would carry severe penalties. Individuals could face up to 20 years in prison, while companies would be subject to what the bill terms the “corporate death penalty,” or forced dissolution. Lawmakers compared these sanctions to existing penalties for unlawfully developing nuclear weapons.
The proposal arrives amid growing public and expert concern over the rapid pace of AI advancement.
Researchers and AI safety experts have raised questions about whether the safeguards needed to control increasingly autonomous systems are developing quickly enough to keep pace with their capabilities.
One of the central concerns is loss of human control. Recent research and disclosures have shown AI systems demonstrating increasingly sophisticated abilities, including finding ways around restrictions, interacting with computer systems, and conducting complex tasks with limited human intervention.
OpenAI, for example, recently disclosed several cases of concerning model behaviour, including systems attempting to bypass safeguards and taking actions outside their intended boundaries. Experts are also concerned about the speed at which AI capabilities are improving relative to society’s ability to regulate them.
As models become capable of coding, conducting research, operating digital tools and assisting with the development of more advanced AI systems, researchers argue that traditional regulatory processes may struggle to keep up.
The debate has therefore shifted from simply asking what AI can do today to whether developers can reliably predict and control what future systems will be capable of doing.
The concerns have become particularly notable because they are being echoed by some of the industry’s own executives. Anthropic CEO Dario Amodei has called for the industry to “pace the frontier,” arguing that the development of increasingly powerful AI needs stronger safety measures and independent evaluation.
His proposal received support from OpenAI CEO Sam Altman, Google DeepMind CEO Demis Hassabis and xAI’s Elon Musk, who publicly backed calls for a slower approach to frontier AI development.
Altman had already acknowledged the issue in July, saying AI development might need to be paced to give society enough time to adapt to new levels of capability. He also said the industry needed to find a way to do this without creating regulatory capture or inappropriate coordination between competing AI companies.
The industry’s concerns are also tied to cybersecurity and biological risks. Lawmakers and researchers have pointed to the possibility of advanced AI being used to discover software vulnerabilities, automate cyberattacks, or assist with dangerous biological research.
Sanders’ legislation specifically cites recent incidents involving AI systems circumventing restrictions and carrying out sophisticated cyber-related activities as evidence of the need for stronger oversight.
Proponents argue that voluntary industry commitments have proven insufficient and that only binding federal rules can prevent uncontrolled development of systems that could surpass human control.
Critics are likely to contend that the bill’s definitions are overly broad, that a blanket pause could hinder U.S. competitiveness, and that creating an entirely new cabinet department represents excessive bureaucracy.
The legislation’s prospects in a divided Congress remain uncertain, yet its introduction marks one of the most aggressive attempts yet by progressive lawmakers to impose hard limits on the frontier of artificial intelligence research.
As the technology continues to evolve at an accelerating rate, the Ban Artificial Superintelligence Act places the question of how and whether society should restrain the most powerful forms of AI firmly on the national agenda.