The White House unveiled new rules on Thursday requiring U.S. federal agencies to ensure that their artificial intelligence tools do not harm the public. Vice President Kamala Harris emphasized the importance of verifying that AI tools do not endanger the rights and safety of the American people. By December, each agency must have concrete safeguards in place for AI tools used in various applications, from facial recognition screenings at airports to controlling the electric grid or determining mortgages and home insurance.

The policy directive issued by the White House's Office of Management and Budget is part of President Joe Biden's broader AI executive order signed in October. The directive specifically targets AI tools used by government agencies for decisions regarding immigration, housing, child welfare, and other services. Agencies that cannot apply the safeguards must cease using the AI system, unless there are justifiable reasons for doing so.

In addition to the safeguards, federal agencies are required to hire a chief AI officer with the necessary expertise to oversee AI technologies and must make public an inventory of their AI systems with an assessment of potential risks each year. The new requirements aim to strengthen the positive uses of AI by the U.S., according to Shalanda Young, the director of the Office of Management and Budget.

While the rules exempt intelligence agencies and the Department of Defense, which is having a separate debate on the use of autonomous weapons, the overall goal is to ensure that AI is used responsibly to improve government services and access for the American people.