How Congress Fumbled Regulating Artificial Intelligence

Lawmakers from both parties have said for months that artificial intelligence is one of the most pressing issues of our time.

They may have missed their best chance to regulate it.

Nearly 400 bills mentioning AI were introduced this Congress, and lawmakers held dozens of hearings grappling with the risks and benefits of AI. None of this has translated into a major legislative push to regulate AI companies like OpenAI, Nvidia, Google and Amazon, which have only grown more influential in the economy and in the federal government this year.

Sen. Josh Hawley, who has introduced bills aimed at banning AI chatbots for minors and creating safety requirements for AI developers, told NOTUS that if Congress fails to regulate AI now, it might become harder to do so in the future.

“We should regulate social media to protect kids, and we should do the same with AI. But it’s more pressing with AI just because its capability is gonna be something that’s really tremendous,” Hawley said.

Both Democrats and Republicans share this concern. When NOTUS asked if Congress is missing the window to regulate AI, Sen. Amy Klobuchar, who has introduced bills aimed at curtailing AI use in elections and AI safety, had a simple answer: “Yes.”

“We need to do this now. And it can be reasonable, practical regulation,” Klobuchar said. “We have to do something about intellectual property, about the fact that people don’t have rights to their own images. … There’s so many things we can do right now, but the companies are able to stop everything in their tracks.”

Despite the clear appetite for regulation, congressional leaders have let bipartisan working groups fall apart, dragging out progress on issues lawmakers from both parties say they could get a meaningful majority behind.

In the meantime, states like California and New York have grown impatient with Washington and have passed or enacted legislation of their own. That’s been met by some resistance from congressional Republicans, some of whom have tried — and failed — to pass restrictions on states’ ability to enforce guidelines.

Earlier this year, Republicans tried to include a provision to block state-level AI legislation in the GOP reconciliation package. Then, Republican leaders in the House tried to introduce similar language in this year’s must-pass National Defense Authorization Act. Both efforts failed under bipartisan opposition.

It’s been one of the highest-profile battles and one of the only issues with momentum related to AI on Capitol Hill.

“There’s only so much oxygen in the room, and this useless debate over preemption is sucking all that oxygen up,” Chris MacKenzie, a spokesperson for Americans for Responsible Innovation, which supports federal regulation of AI, told NOTUS about efforts to block state-level regulation.

But Congress is doing little else to try to regulate AI. And the pressure against doing so is only going to increase, given the emergence of super PACs that plan to support pro-AI candidates ahead of the 2026 midterm elections. MacKenzie said this environment is distracting Congress from proactively regulating AI.

“The tech industry is now threatening to throw so much money into elections that they have sidelined some of the conversations about proactive tech governance and put preemption on the front burner,” MacKenzie said. “Part of the challenge is the political environment. We’re headed into an election year, and Congress has, really, seven months to legislate before things start shutting down and people head to the campaign — I’m not writing the year off, but it does become more challenging to move new bills in an election year.”

Proponents of AI regulation on Capitol Hill also see industry money as one of the main obstacles to regulation.

“AI has a tremendous amount of money. It’s almost all the same people as social media; it’s all the same corporations. And they’re hugely influential, and they don’t want anything done,” Hawley said.

However, as more people use AI and the effects of the technology become more obvious, some lawmakers expect AI to become a salient issue among voters.

“I don’t think the average congressperson understands how much the public is scared and angry about AI — and that’s not going away,” Democratic Rep. Bill Foster, also a former member of a House AI bipartisan task force, told NOTUS. “I wouldn’t be surprised if by the next election one of the main issues here would be what is being done about AI.”

“The idea that there’s this data center in my back yard that is doubling the price of electricity, or that AI is taking my jobs and that the government should do nothing about it is going to be a massive loser,” Foster said. “The thinking is always short-term, and that’s the problem with politics.”

Last year, both House and Senate leaders created bipartisan working groups to draft AI policy. Efforts like the House Bipartisan Task Force on AI and the Senate AI Working Group produced dozens of policy recommendations to regulate AI.

Congressional Republican leaders did not continue these dedicated working groups this year, and only one of the recommendations in those reports has made it to President Donald Trump’s desk: a law that allows victims whose likeness is used in AI porn without their consent to request their image to be taken down from the internet.

Most other AI uses remain unregulated.

Neil Chilson, an AI policy researcher at the right-leaning Abundance Institute who supports blocking state-level regulation of AI, said losing these working groups is bad for developing a national regulatory standard.

“AI isn’t just one thing; It’s this big, massive, general-purpose technology. And that means, at least procedurally in Congress, there are aspects of it that touch every single committee jurisdiction,” Chilson told NOTUS. “These working groups are valuable to hash that out procedurally.”

“Like, how does the fact that this is a general-purpose technology and that there’s going to be aspects of it that touch every single part of the economy play out in the legislative process?” Chilson said. “Being more familiar with that and building some muscle memory about who owns what issues is really important.”

Some Republicans in Congress insist there’s still a bipartisan coalition that is willing to work on AI regulation.

“There’s just as much bipartisan appetite to pass legislation this year as last year,” California Republican Rep. Jay Obernolte, former co-chair of the House Bipartisan Task Force on AI, told NOTUS.

“What you’ve seen is that the effort is getting caught up in the partisan polarization that you’re seeing in politics and that everything else in Washington has gotten caught up in the last nine months,” Obernolte said. “There’s an awful lot of substance and technology policy that the two parties broadly agree on.”