AI's growing role in rulemaking raises new transparency questions

While lawmakers are debating what an appropriate regulatory framework for artificial intelligence should be, experts who spoke at a panel on Monday noted that federal employees have been using the technology for years to develop all sorts of regulations. 

Cary Coglianese, a regulation professor at the University of Pennsylvania, said that agencies, for example, have been using AI to analyze thousands of public comments that some proposed rules receive. 

“Back in the day, contractors would literally photocopy comments, clip them out and sort them in different categories — and do what essentially natural language processing tools can do now so quickly and rather effectively,” he said. 

But as the technology advances, speakers on the panel — which was sponsored by the Coalition for Sensible Safeguards, a left-leaning group of more than 220 regulatory-focused organizations — argued that AI could take over more steps in the rulemaking process, potentially supercharging the Trump administration’s efforts to slash regulations and downsize the federal government. 

“I worry a great deal that when you have a government and policymakers basically out to destroy huge portions of the public sector, wholesale implementation of AI will provide them a social veneer just to do that rather than using technology in more positive ways to improve the quality and breadth of public services,” said Ben Zipperer, a senior economist at the Economic Policy Institute. 

There are higher stakes for federal agencies compared with private companies when it comes to AI implementation, Zipperer contended, because improper usage could lead to individuals losing their benefits or shaky legal justifications for government actions. Unlike in the private sector, however, the economist emphasized that there isn’t a market factor in government to push back on poor AI implementation. 

“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 said. “So if the public sector uses AI as a way or an excuse to make public servants' jobs worse or to cut public services, there's not a non-political check on that, and public services just suffer.” 

There already are high-profile examples of the Trump administration using AI in its cost-cutting initiatives. Department of Government Efficiency staffers in 2025 prompted ChatGPT to assess National Endowment for the Humanities grants and respond in fewer than 120 characters whether they fall under diversity, equity and inclusion programs that officials were targeting for elimination. 

Based on that incident, which came to light due to a lawsuit over the canceled grants, Jonathan Walter, a senior policy counsel at the Leadership Conference on Civil and Human Rights, warned that officials could use AI to shut out public feedback they object to on proposed rules. 

“It would have been just as easy for them to say, ‘Hey, we're just going to delete or ignore any comment with any mention of race or disparate impact or gender or whatever protected class might have a stake in a rulemaking,’” he said. “Then, there are all of the different downstream consequences that stem from that. It could be that the final rule happens to ignore certain anti-discrimination laws. It might not reflect the amount of participation from a certain group.” 

J.B. Branch, a technology expert at Public Citizen, added that he’s worried about which AI systems feds may use — specifically calling out Grok, which has been criticized for generating racist comments as well as responses that are aligned with the opinions of its creator, former DOGE head Elon Musk. 

“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,” he said. 

Last year, the president signed an executive order that directed agencies to procure “ideologically neutral” and “truth-seeking” AI that is not “woke.” 

The Trump administration is pursuing many deregulatory actions, with officials writing in their 2026 regulatory plan that they will “achieve extraordinary cost savings for Americans” and “promote liberty, unleash American energy dominance, preserve products consumers love and eradicate the ideology of [DEI].” 

But Lauren McFerran, the executive director of the AFL-CIO Tech Institute, cautioned that an overreliance on AI during the deregulatory process could result in increased judicial scrutiny rather than more rules being cut. 

“A court is going to be looking at whether [the agency has] made a rational and supportable connection between the evidence and their reasoning and the ultimate decision. When you get to the more advanced [AI] systems, they're learning from their own processes,” she said. “That creates — the term that tends to be used is a black box. You don't really necessarily know what steps the product undertook to get to certain outcomes, and I think that could be potentially problematic for a court that's trying to review whether the agency, in fact, engaged in the reasonable decision-making process that it was supposed to.”