The Biden administration is set to enforce a new regulation mandating developers of major artificial intelligence (AI) systems to disclose safety test results to the government.
The White House AI Council will convene on Monday to assess the progress made on President Biden's executive order, signed three months ago, aimed at effectively managing the rapidly evolving AI technology landscape.
A key aspect of the executive order, within the 90-day timeframe, is the requirement under the Defense Production Act for AI companies to share crucial information, including safety tests, with the Commerce Department.
Ben Buchanan, White House special adviser on AI, emphasized the importance of ensuring AI systems' safety before public release, stating that companies need to meet this standard set by the president.
While software companies have committed to conducting safety tests, there is currently no common standard for these tests. The National Institute of Standards and Technology will develop a unified framework for assessing safety, as outlined in Biden's October order.
AI has emerged as a top economic and national security priority for the federal government, given the significant investments and uncertainties surrounding new AI tools like ChatGPT, which can generate text, images, and sounds. The administration is also engaging in legislative efforts and international collaborations to establish regulations for managing AI technology.
The Commerce Department has drafted a rule concerning U.S. cloud companies that provide servers to foreign AI developers, reflecting efforts to address potential security concerns.
Nine federal agencies, including Defense, Transportation, Treasury, and Health and Human Services, have conducted risk assessments regarding AI's impact on critical national infrastructure, such as the electric grid.
Additionally, the government has intensified the recruitment of AI experts and data scientists within federal agencies to enhance readiness in managing this transformative technology.
Buchanan emphasized that while AI holds immense transformative potential, regulatory measures are necessary to ensure its responsible and effective deployment.
