Rep. Lori Trahan’s decision last week to forgo a leadership bid to instead focus on developing artificial intelligence policy surprised some of her colleagues. It also left some of them wondering whether her thwarted attempt to work across the aisle on the subject was what doomed her prospects.
In an interview Friday, the Massachusetts Democrat brushed off the idea that efforts to broker an AI deal with a Republican backfired. She said the opposite was true: Many House Democrats told her she should run for caucus vice chair because of her work in this complex policy space.
“Look, I spent August really thinking it through, and it came down to a simple question that I think a lot of people don’t ask themselves enough, and that’s just: What I’m most passionate about?” Trahan recalled.
It turned out that the chance to help Democrats formulate a vision for how to regulate a rapidly evolving industry was what she couldn’t pass up.
“When I commit to something,” she said, “I go all in.”
There’s no guarantee Trahan will be successful, with members of both parties reluctant to work together and slamming each other over the issue on the campaign trail. But Trahan isn’t retreating or shying away from her work with Rep. Jay Obernolte (R-Calif.).
She spoke with POLITICO about how she sees the AI landscape from a political and policy perspective, capping off a week of major developments in Washington — including the introduction of an aggressive plan to pause development of superintelligence models, an escalation in Democratic attacks on Republicans for the rise of data centers, President Donald Trump’s vow to “Let Data Reign,” and new reporting about a tech industry giant’s attempts to talk Trump out of onerous regulations.
This interview has been edited for length and clarity.
House Democratic Leader Hakeem Jeffries came out swinging this past week in trying to pin the proliferation of data centers on Republicans. What do you think of Democratic messaging on this?
I do think that the Republicans are having trouble reconciling what the president is saying with what their constituents are demanding, and that, I’ll just say, people aren’t imagining this. They’re already feeling it … in their electric bill. They see it when a new data center gets proposed down the road.
They’re obviously reading headlines about AI models that break out of their own labs, and I think those worries are legitimate. So I don’t think the folks are necessarily antitechnology, but they want to know that somebody’s watching this — that the most powerful technology in human history isn’t running on the honor system, and right now, it basically is.
Congressional Progressive Caucus Chair Greg Casar has made a huge push for Democrats to be, as he put it, “AI populists.” What do you think of progressives trying to carve out a lane for themselves on AI, and what has your work with them on this issue been like?
I think for a long time, the hardest part of this job was convincing people that this is an issue that couldn’t wait. I think that part is largely over, right? AI is advancing at a staggering pace, and the risks are scaling right alongside the capabilities. So I think after a summer of breaches … members on both sides, they’re just done waiting, and I think you’re seeing a wave of proposals … I think that just reflects a new sense of urgency in Congress that I couldn’t be happier to see.
Casar and Sen. Bernie Sanders announced a plan to ban the development of AI “superintelligence” and pause work on cutting-edge models until an agency is stood up to monitor risk. What is your reaction to the proposal?
My team and I are still reviewing the bill, obviously gathering feedback from the safety community, so I’m not going to prejudge the details. Of course, I share the concern driving it. I mean, these companies are losing control of their own models, and they’ve said so themselves.
I’m always a little cautious about outright bans, moratoriums, or blanket pauses because you really only get one shot at something like that, and if you time it wrong, you can end up worse off. … You trigger a backlash that chokes off the chance to put sensible, durable guardrails in place in the future.
A permanent ban with no path to passage, that doesn’t protect anyone. … But right now I get most nervous that we’re not passing the bill that can pass, that matches sort of where we are right now, which is, models escaping their environment and us relying on the honor system from these companies.
You are referring to your bill with Obernolte, which would establish a national framework for regulating advanced AI models. What’s happening with it? You were pushing for committee action before the midterms. That seems all but impossible now.
The calendar isn’t on our side before the midterms … so I have shifted my focus to before the end of the year. I think that’s the realistic window to get this done this year. Jay and I are working our respective sides of the aisle. … I talked to a lot of rank-and-file members on both sides of the aisle, and they’re all scratching their heads, wondering why we’re not taking this up.
So why do you think it hasn’t moved?
I don’t sit in the minds or the rooms of Republican leadership. … I am worried that they are comfortable with the president sort of managing this through executive fiat rather than putting something in place that is durable and predictable.
I’m hoping that with internal pressure and with external pressure, that the leadership, Speaker [Mike] Johnson and Leader [Steve] Scalise, will take this up and debate it at a minimum.
What is your reaction to the phone call our colleagues reported on, where Meta CEO Mark Zuckerberg urged Trump to scrap a plan to stand up a federal AI regulator?
I think that there are a lot of members of Congress who have ideas in terms of how this gets regulated.
Reasonable people can disagree on what regulatory construct we come up with. What I don’t think the American people agree with is that Mark Zuckerberg and President Trump are going to figure that out on a phone call.
Democrats have been fairly scattered on AI policy this Congress, and that could make it difficult for the party to unify around a policy platform while still negotiating with Republicans. How are you thinking about the next Congress?
I’m excited about next Congress just because I know when Democrats are in the majority, we’re going to get to work. … I think the fact that Democrats are talking about it a ton and we’re not under the thumb of the leader of our party, so that we can’t say or reflect any of the concerns that we hear back at home, it’s how it should be.
The fact that we’re able to have those conversations and we’re all coming from diverse places is a good thing. I think that’s what the American people want to see. They want to see us engage in every corner of the country in terms of concerns around AI and have a debate about it.
That’s not happening right now. Right now, we’re mostly just hearing from the president having a phone call with a rich tech executive or using export controls at his discretion.
Read the full article here