Biden Executive Order Mandates Safety Testing for Powerful AI Models
President Biden issues a new executive order requiring developers of powerful AI systems to share safety test results with the federal government before public release. The directive aligns with the Defense Production Act to address national security and public health risks.
President Biden issues an executive order that establishes new standards for artificial intelligence safety and security. The directive requires companies developing the most powerful foundation AI models to notify the federal government and share all safety test results before deploying their systems to the public.
This order targets any foundation model that poses a potential risk to national security, economic security, or public health by aligning these requirements with the Defense Production Act. The administration states that these measures ensure AI systems remain safe, secure, and trustworthy before companies release them to consumers.
The announcement arrives alongside a global push for AI governance, including a voluntary code of conduct from G7 leaders, a new United Nations AI governance board, and a global safety summit in the U.K. Furthermore, the National Institute of Standards and Technology receives a mandate to develop additional tools and systems to guarantee AI trustworthiness.