Federal agencies’ use of AI in rulemaking raises transparency and bias concerns

Federal agencies have used artificial intelligence for years to process public comments and assist in drafting regulations, but experts warn the technology may now supercharge a deregulatory push by making agency decisions harder to explain and defend.

At a panel hosted by the Coalition for Sensible Safeguards, Cary Coglianese, a regulation professor at the University of Pennsylvania, noted that agencies once relied on contractors to manually sort public comments, a task now handled quickly and effectively by AI tools. However, he and other panelists cautioned that AI’s expanding role in rulemaking could obscure decision-making processes as agencies cut regulations and downsize government operations.

Ben Zipperer, a senior economist at the Economic Policy Institute, argued that unlike private companies, federal agencies lack a market-based check on poor AI implementation. He said, “A private sector company could replace all their workers with AI, but then their product quality would probably suffer, and they'll lose sales. But there's no profitability check on the public sector.” He warned that AI could be used to justify cuts to public services without non-political oversight.

A 2025 incident highlighted these risks when Department of Government Efficiency staffers used ChatGPT to assess National Endowment for the Humanities grants, responding in fewer than 120 characters whether they fell under diversity, equity, and inclusion programs targeted for elimination. The incident, revealed through a lawsuit over canceled grants, raised concerns that AI could be used to suppress public feedback on proposed rules.

Jonathan Walter, senior policy counsel at the Leadership Conference on Civil and Human Rights, warned that officials could use AI to ignore public comments referencing protected classes, such as race or gender, potentially leading to rules that ignore anti-discrimination laws or underrepresent certain groups.

J.B. Branch, a technology expert at Public Citizen, raised concerns about specific AI systems used by federal agencies, citing Grok’s history of generating biased or ideologically aligned responses. He said, “If someone in the federal government is using Grok, there is a deep concern that whatever output it's providing could be a potentially very biased output.”

Last year, the president signed an executive order directing agencies to procure “ideologically neutral” and “truth-seeking” AI. However, panelists noted that the administration’s 2026 regulatory plan emphasizes cost savings and deregulation, including targeting diversity, equity, and inclusion initiatives.

Lauren McFerran, executive director of the AFL-CIO Tech Institute, cautioned that overreliance on AI in deregulatory processes could invite increased judicial scrutiny rather than achieving the administration’s goal of cutting more rules.