As artificial intelligence grows more powerful and autonomous, policymakers are increasingly debating whether limits should be placed on how far the technology can advance.
In the United States, Senator Bernie Sanders and Representative Greg Casar have introduced legislation that would prohibit the development and deployment of artificial superintelligence (ASI) while imposing a temporary pause on certain advanced AI development until stronger safety standards are established. The proposal also calls for the creation of a federal agency dedicated to AI safety and would direct the United States to pursue international agreements aimed at preventing the development of ASI.
What is artificial superintelligence?
ASI refers to a hypothetical form of AI capable of outperforming humans across virtually every intellectual task. While today’s AI systems can already write code, analyze information, generate content, and operate digital tools, ASI would represent a significantly more advanced level of capability, with potentially far greater risks.
The U.S. proposal is not an isolated effort. Other lawmakers are focusing on regulating autonomous or “rogue” AI systems rather than banning advanced AI altogether. Meanwhile, California has been exploring stricter safety requirements, including independent testing and emergency shutdown mechanisms for advanced AI systems.
The debate has gained urgency as AI systems increasingly demonstrate the ability to carry out complex, multi-step tasks with minimal human oversight. Supporters of stricter restrictions argue that growing AI autonomy could create serious cybersecurity, economic, and national security risks. Critics, however, argue that governments should focus on regulating and rigorously testing powerful AI systems rather than prohibiting future technologies altogether.
For now, the proposed U.S. legislation remains a bill, not law. Its introduction is nevertheless significant because it places a consequential question at the center of the political debate: Should society regulate extremely powerful AI, or seek to prevent certain forms of it from being developed altogether?
As AI continues to advance, governments, technology companies, and researchers will face increasing pressure to balance technological innovation with safety, accountability, and meaningful human control.