Sen. Bernie Sanders and Rep. Greg Casar have introduced the Ban Artificial Superintelligence Act, which would criminalise the development of systems capable of destroying humanity or overthrowing the government. The legislation mandates up to 20 years in prison for any individual or entity that breaches this prohibition. It also requires a pause on advanced AI development while a new federal agency oversees safety research. President Donald Trump has previously suggested the US will adopt the term superintelligence in official documents to replace the broader label of artificial intelligence. The bill defines the technology strictly by its potential for catastrophic harm rather than by current computational capabilities.
This proposal highlights the growing political concern that current safety measures are insufficient to prevent existential risks from emerging technologies. Lawmakers aim to shift the regulatory burden onto developers through severe penalties rather than relying solely on voluntary industry standards. The inclusion of a development pause suggests a preference for caution over speed in the early stages of this technology’s growth.
* The bill creates a specific federal agency to monitor and enforce these restrictions.
* Prison sentences could reach 20 years for first-time violations of the act.
* The definition of superintelligence focuses on the potential for destruction rather than specific metrics.
