View / The AI industry balances Trump and a changing Washington

The tech industry is facing strange headwinds in Washington after meeting with President Donald Trump and senior administration officials this week and signing a “morally binding” accord on AI.

The meeting with Trump was largely successful for all of the companies involved. But now these same firms have to navigate maintaining ties with the president while placating Democrats who could very well control both houses of Congress next year. That’s not to mention what could be a massively consequential midterm election, fueled by AI anxiety among voters, that could test the bounds of their power.

“Companies are going to have to make it more specific, on how they’re going to operationalize it,” Chris Lehane, OpenAI’s chief global affairs officer, told Semafor about the new White House accord on “Super Intelligence” signed by the leaders of several top firms and Trump this week, including OpenAI President Greg Brockman.

“We’ve certainly approached this believing … it’s also in our self-interest to make sure that people feel that we are actually building things safely and deploying responsibly,” Lehane said, adding that OpenAI’s support for certain state-level AI bills and its hopes for a federal bill will remain the same.

“That’s almost either a democratic market force, or the actual market forces themselves. Like, if you’re making cars, you’re going to want to make sure the cars are safe, or people aren’t going to buy your cars, whether there are regulations or not,” he added.

Before lawmakers departed Washington to focus on campaigning ahead of the November elections, AI companies were still in active discussions on Capitol Hill. They’re trying to influence forthcoming bills and determine where they can find consensus with lawmakers, a tech industry source and a source who works on AI policy told Semafor.

Companies know that the midterms are likely to mean a Democratic-run House and possibly a Democratic-run Senate, and with that will come more momentum to pass AI legislation — and likely more scrutiny of their operations, particularly for firms that have demonstrated ties with the Trump administration. “There’s a world where people are smarter and smarter about these issues, and bills are going to be introduced imminently that are much better than what we’ve seen,” the AI policy source told Semafor.

After the election, legislating is an obvious place to go, given the public and many lawmakers in both parties want to see something done, the tech industry source told Semafor.

Considering where the Trump administration was a year ago — that is, attempting to ban state-level AI bills — progress toward federal AI policy is slow, but it’s inching forward, the tech industry source said. The FTC also kicked off its own AI inquiry, which may not yield much in the end, but nevertheless represents a signal from the administration that regulators are prepared to use existing laws to go after any companies that they suspect have hurt consumers.

Still, the White House’s pro-AI posture is unlikely to fundamentally change, and Trump will remain the most important voice on it until the end of his term. Groups that are pushing for data center expansion and industry-friendly laws have plenty of money and resources, plus their own complicated financial ties to the AI and tech companies themselves.

Anthropic may be experiencing the most whiplash since the surreal AI summit: Days after a federal court upheld its designation by the Pentagon as a supply chain risk, CEO Dario Amodei dined with Trump at the White House.

The administration has repeatedly painted Anthropic as the face of a pro-AI safety movement that is “woke” and composed of “doomers” — too focused on AI’s harms and trying to gin up fear for ulterior motives.

“I think they’re going to try to walk this tightrope for two years,” the person who works on AI policy told Semafor. “They’re in a weird place. [Undersecretary of Defense] Emil Michael hates you, and is tweeting about you every day. Nonetheless, the government seems to find you indispensable to their operations.”

Sarah Heck, a policy lead at Anthropic, posted on X following the White House meeting: “You can’t do safety from second place.” She shared a video clip of Amodei applauding the accord while saying AI still has very real risks and how companies address them is still under discussion.

On Thursday, Heck added: “To be clear, I’m talking about the US and its allies here: We can’t ensure AI is developed safely unless the US is in the lead.”