What AI companies want for the millions they’re spending on elections

AI executives and companies believe spending millions in the 2026 midterm elections will allow them to influence AI bills being developed in Congress.

As of the end of June, the two largest AI political action committees had invested at least $44 million in 40 House and Senate candidates, according to CNBC’s analysis of Federal Election Commission data. It’s an early look at how the groups will spend the more than $200 million they raised over the remainder of the primary season and the general election, based on fundraising totals provided by the groups.

The nascent AI industry’s spending is making it an increasingly powerful player in the Washington sphere of influence. Through their PACs, companies are preparing to shape what will be the first national legislation to regulate the use of artificial intelligence.

Brad Carson, who heads Public First Action, a nonprofit with many PACs, said he is seeing more bills being introduced and debated. Concerns are rising about AI legislation, especially as concerns about the capabilities and risks of powerful AI models such as Mythos and Claude Fable come to the fore. While it’s unlikely any legislation will cross the finish line this year, given the limited number of days lawmakers are in session, both parties have signaled that artificial intelligence will remain a priority in the coming years.

“They have a lot of benefits. They have a lot of dangers. And you can’t release them into the wild without government concern,” Carson told CNBC. “From right to left, from pro-Trump to anti-Trump, everyone knows this.”

Finding the right regulatory structure is critical for lawmakers, said Josh Vlasto, co-chair of Leading the Future.

“It’s very important that we do this now and urgently, because this is still in the early stages of the technology but it’s being adopted quickly and at scale,” he told CNBC.

Multiple candidates supported by both PACs won the primaries. Of the 28 leading candidates, 25 have won the primaries, two have not yet made it to the ballot, and only one – Jesse Jackson Jr. – lost. The group also opposed Alex Bores, who lost the Democratic primary in New York’s 12th Congressional District.

Public First Action supported candidates in 11 races. All the candidates he supported won, except Bores. Carson said the group plans to spend on 50 to 60 races by the end of the midterms.

The playbook used by artificial intelligence companies is not new. In the 2024 elections, crypto-backed PAC Fairshake supported pro-crypto candidates on both sides of the aisle, investing an eye-popping $200 million into the election. Result: A major bill on stablecoins has been signed into law, making significant progress on the standard digital assets bill favored by major crypto companies. coinbase and Ripple.

Leading the Future spent more than $24 million on primaries through the end of June, according to data filed with the Federal Election Commission. The group said it has raised $125 million by the end of 2025, in part from private equity firm Andreessen Horowitz and donors including Open AI co-founder Greg Brockman. palantir co-founder Joe Lonsdale, SV Angel founder Ron Conway, and artificial intelligence software company Perplexity.

Launched last year, Public First Action has spent $20 million so far and announced last month that it had raised $80 million by the end of June. According to a PAC spokesperson, the group received $20 million from Anthropic; However, this amount is not limited to political purposes, but to educate the public on artificial intelligence policy.

Pubic First Action does not disclose its donors; Antropik announced his own donation. But Carson said the group has received donations from OpenAI, Google, DeepMind and X employees.

The two groups have faced off numerous times in races, spent against each other in the Manhattan Democratic primary and even traded barbs in interviews.

But the policy differences between the two are much more nuanced than “pro” or “anti” regulation. Both groups support guardrails to some degree and even overlap in areas such as the need to protect children online.

The biggest differences touch on one of the thorniest issues in Washington: Whether a single federal standard should trump state laws on artificial intelligence. But even on this issue, the groups are not on completely opposite sides of the debate.

Vlasto said in an interview with CNBC that Leading the Future advocates for “a broad, national, consistent framework for regulation governing artificial intelligence.” He denied that the group violated state law and pointed to his support of New York’s groundbreaking artificial intelligence law, the RAISE Act., Bores helped lead the charge as a New York assemblyman.

But the RAISE Act shows how complex the group’s stance is. Leading the Future spent nearly $8 million opposing Bores; this was due to his push for a more aggressive UPGRADE Act than was eventually enacted.

Before the bill was signed into law, New York Governor Kathy Hochul successfully pressured lawmakers to pass changes that would weaken AI companies’ reporting requirements and the size of penalties; this brought New York law more in line with the law in California. These changes led Leading the Future to support the final legislation and still oppose one of the legislators who supported the earlier version of the bill.

Public First Action further supports state laws and fights efforts to block them; but Carson said that if Washington “can find a comprehensive federal approach to these problems, then prevention will become a natural part of our constitutional order.”

Republicans on Capitol Hill have tried and failed several times to block state legislation. House Majority Leader Steve Scalise, R-La., told CNBC that state laws “hurt innovation” and that overriding them “will be the foundation of everything we do.”