WASHINGTON (Reuters) – The White House said President Donald Trump is signing an executive order on Thursday setting guidance for federal agency use of artificial intelligence that aims to foster public acceptance of the technology in government decision making.
The order directs agencies to prepare inventories of AI-use cases throughout their departments and directs the White House to develop a roadmap for policy guidance for administrative use of AI.
Michael Kratsios, U.S. chief technology officer, said the executive order “will foster public trust in the technology, drive government modernization, and further demonstrate America’s leadership in artificial intelligence.”
The Trump administration has made artificial intelligence a priority, earlier issuing guidance to federal agencies aimed at limiting “overreach” in regulating the use of AI by private companies, while urging agencies to use AI to eliminate outdated regulations.
The order emphasizes AI use must be “lawful; purposeful and performance-driven; accurate, reliable, and effective; safe, secure, and resilient; understandable; responsible and traceable; regularly monitored; transparent; and accountable.”
AI is already used by numerous government agencies for predictive enforcement tools and by regulatory agencies to process and review vast amounts of data to detect trends and shape policymaking.
Some U.S. states and cities have raised concerns about AI applications, especially possible algorithmic bias in the use of facial-recognition software by law enforcement.
A February report by Stanford and New York University researchers submitted to a U.S. administrative agency documented 157 use cases of AI by 64 federal agencies and said it could “modernize public administration, promoting more efficient, accurate, and equitable forms of state action.”
But the report titled “Government by Algorithm” noted AI can be used to deny benefits or make decisions affecting the public’s rights and could boost concerns about arbitrary government action.
“A crucial question will be how to subject such tools to meaningful accountability,” the report said.
Reporting by David Shepardson; Editing by Aurora Ellis