Politicians are slowly but surely starting to try out AI for themselves

• More and more lawmakers are beginning to use AI themselves.
• That includes some who told Business Insider just months ago that they hadn't gotten into it.
• "I use it, despite the fact that I think it's going to destroy us," said one senator.

As recently as June, Sen. Elizabeth Warren was one of several lawmakers who resisted using AI, owing to a skepticism of the technology's ability to deliver accurate information.

But now?

"Yeah, that's changed," the Massachusetts Democrat said with a laugh this month, explaining that she now finds ChatGPT to be "really valuable" for basic research questions, even if she still catches the occasional hallucination.

Warren said that she began using ChatGPT more after seeing her daughter use it. She says she doesn't "rely" on the technology, but uses it to "start to approach a problem."

"Like, I'm in the middle of reading something, and I think: How many people are there in Mississippi? And what's the breakdown of little children, and people over 65?" Warren said. "I pop that into ChatGPT and get an answer that's better than a straight Google answer. I can get more detail, and more ways to slice and dice the numbers."

Other skeptics are tiptoeing their way into using the technology, including some who are otherwise major critics of the AI industry.

"The other day, I decided, you know what, maybe after doing all these hearings on ChatGPT, I should at least see how it works," Republican Sen. Josh Hawley of Missouri told Business Insider, saying he asked a "very nerdy historical question" to ChatGPT about the Puritans in the 1630s. "I will say that it returned a lot of good information."

"I use it, despite the fact that I think it's going to destroy us," Democratic Sen. Chris Murphy of Connecticut told Business Insider.

At the highest levels of the American government, personal AI adoption remains spotty. White House Press Secretary Karoline Leavitt told reporters in November that she doesn't think President Donald Trump uses the technology himself.

"I've never witnessed it," she said. "So I can't attest to that."

Vice President JD Vance, meanwhile, declared himself a "Grok guy" in an interview with Fox News's Sean Hannity in November.

"I think it's the best," Vance said of the Elon Musk-owned AI chatbot. "It's also the least woke."

But House Speaker Mike Johnson said that he hasn't used AI himself, saying on the Katie Miller podcast that he simply hasn't had the "luxury of time" to get into it.

"I just haven't gotten into it yet. My life is not normal right now, okay?" Johnson said. "And AI has really been — it's become in popular use really during the term of my speakerhip, for the last two years. And so I just haven't had time to engage."

As lawmakers have begun working more with AI, some have had strange experiences. Democratic Rep. Jared Huffman of California told Business Insider that at one point, he tried to use Microsoft Copilot to look up what was inscribed on the bullet casings of Charlie Kirk's alleged murder suspect.

Instead, Huffman says he ended up in an argument, and that Copilot insisted that Kirk was still alive despite the congressman prompting the AI with information to the contrary.

"It continued to fight with me, insisting that the whole assassination was a conspiracy theory," Huffman said. "It was freaking weird."

Most lawmakers who spoke with Business Insider said that they use ChatGPT.

But at least one — Democratic Rep. Don Beyer of Virginia — said he preferred to use Anthropic's Claude, owing to the company's focus on safety and ethics in AI.

"I was very impressed with the fact that they actually put together their own Constitution on the ethical use," Beyer said. "They seem to be — at least they're positioned as — more enlightened."