Tech experts say federal government needs more AI leadership positions at agencies
Federal agencies are lacking critical points of contact for the government to coordinate on responsible oversight and use of artificial intelligence tools, experts said Tuesday.
Government officials and technology experts testified Tuesday that the federal government is lacking adequate oversight for its use of artificial intelligence tools and called on agencies to appoint chief AI officers to safeguard against potential risks.
Lynne Parker, director of the University of Tennessee’s AI Tennessee Initiative and former director of the National Artificial Intelligence Initiative Office, told the Senate Homeland Security and Governmental Affairs Committee that establishing a chief AI officers council would help the government "coordinate in a consistent way" across all agencies.
"We don't have a single point of contact or a single responsible person at the agencies that oversee these activities," said Parker.
Under Executive Order 13960, agencies are required to identify points of contact around their use of AI tools and make their AI use case inventories public on accessible digital databases. However, while some agencies – like the departments of Health and Human Services and Defense – have brought on inaugural chief AI officers, others have yet to make the necessary appointments.
The 2020 executive order also requires agencies to ensure their use of AI applications are safe, secure and resilient against systematic vulnerabilities and malicious exploitation.
But cybersecurity experts and technologists have recently warned about a policy vacuum when it comes to AI tools, while urging lawmakers to develop more regulatory guidance for agencies and organizations that are already leveraging many AI software products and technologies.
Parker added that the council could be led by the Office of Management and Budget and the Office of Science and Technology Policy's National AI Initiative Office, with representation from the General Services Administration's Center of Excellence in AI, to "provide the expertise across the federal agencies to coordinate these processes and provide leadership for the government as a whole."
The experts testified that collaboration was a critical component in developing adequate AI oversight and regulations for the public and private sectors – including overseas partnerships with U.S. allies to preserve American interests in the field of emerging tech.
Daniel Ho, a Stanford Law School professor, recommended establishing a multilateral AI research institute "to enable like-minded countries to collaborate together" on creating standards around responsible AI use.
The federal government's issues with adequate AI oversight extend beyond identifying points of contact across agencies, according to Taka Ariga, chief data scientist for the Government Accountability Office. Ariga called on Congress to establish a digital services academy for the federal government to improve the pipeline of trained cyber talent that can work on various modernization projects throughout the government, "including the implementation of AI systems."
"The federal government as a whole continues to face barriers in hiring, managing and retaining staff with advanced technical skills — the very skills needed to design, develop, deploy and monitor AI systems," said Ariga.