AI executives and companies are betting that by spending millions of dollars in the 2026 midterm elections, they can influence AI legislation being developed in Congress.
As of the end of June, the two largest AI political action committees had spent at least $44 million on 40 House and Senate candidates, according to a CNBC analysis of Federal Election Commission data. This is an early taste of how the groups will spend the more than $200 million they have raised for the remainder of the primary season and the general election, according to fundraising totals provided by the groups.
Spending by the burgeoning AI industry has made it an increasingly powerful player in Washington’s sphere of influence. Through PAC, companies are gearing up for how they could shape the first national law regulating the use of AI.
Brad Carson, who heads Public First Action, a nonprofit organization with multiple PACs, said there has been an increase in the number of AI bills being introduced and debated, especially as concerns about the capabilities and risks of powerful AI models like Mythos and Claude Fable are in the spotlight. Given the limited number of days in Congress, it is unlikely that any legislation will cross the finish line this year, but both parties have signaled that AI remains a priority going forward.
“They have a lot of benefits, but they also have a lot of risks, and we can’t ignore the government’s concerns and release them into the wild,” Carson told CNBC. “Everyone recognizes that, from the right to the left, from pro-Trump to anti-Trump.”
Josh Vlasto, co-leader of Leading the Future, said it was important for lawmakers to come up with the right regulatory structure.
“It’s very important to do this now and urgently because this technology is still in its infancy, but it’s being deployed quickly and at scale,” he told CNBC.
An overwhelming number of candidates supported by both PACs won their primaries. Of the 28 candidates that Leading the Future supported, 25 won their primaries, two have not yet entered the race, and only one, Jesse Jackson Jr., lost. The group also opposed Alex Boaz, who lost the Democratic primary for New York’s 12th Congressional District.
Public First Action is endorsing candidates in 11 races. All candidates supported by the company won, except for Mr. Boas. Carson said the group plans to participate in 50 to 60 races by the end of the midterm elections.
The strategies AI companies are using are not new. In the 2024 election, the crypto-backed PAC Fairshake spent an eye-popping $200 million on the campaign, supporting pro-crypto candidates on both sides of the aisle. The result: Major stablecoin legislation was passed, and significant progress was made on the Rules of the Road Digital Assets Bill, which is supported by major crypto companies including: coinbase And Ripple.
Leading the Future has so far spent more than $24 million in the primary through the end of June, according to data filed with the Federal Election Commission. The organization is led by private equity firm Andreessen Horowitz, OpenAI co-founder Greg Brockman, Palantir Co-founder Joe Lonsdale, SV Angel founder Ron Conway, and AI software company Perplexity.
Public First Action, launched last year, has spent $20 million and announced last month that it had raised $80 million by the end of June. The group received $20 million from Anthropic, but a PAC spokesperson said the money was limited to educating the public about AI policy, not for political purposes.
Public First Action does not disclose its donors. Anthropic revealed its donation. But Carson said the group has received donations from employees of OpenAI, Google, DeepMind and X.
Two AI groups invest millions in politics
The two groups have clashed numerous times on the campaign trail, competing against each other in Manhattan’s Democratic primary and even exchanging swipes in interviews.
But the policy differences between the two are much more subtle than being “for” or “against” regulation. Both groups support some degree of guardrails and overlap in areas such as the need to protect children online.
The biggest difference touches on one of Washington’s most vexing issues: whether a single federal standard should supersede state laws on AI. But even on this issue, the two groups are not completely at odds.
Leading the Future advocates for “a broad, national, consistent framework for regulation governing AI,” Brust said in an interview with CNBC. He denied that the group violated state law and pointed to its support of New York’s landmark AI law, the RAISE Act, which Boas helped lead as a New York state lawmaker.
But the RAISE Act shows how complex the group’s position is. Leading the Future spent nearly $8 million opposing Boas, largely because he pushed for a more aggressive RAISE Act than was ultimately signed into law.
Before the bill was signed into law, New York Governor Kathy Hochul successfully pressured lawmakers to agree to changes that would ease the scale of reporting requirements and penalties for AI companies, bringing New York’s law more in line with California’s. These changes resulted in Leading the Future supporting the final bill, while still opposing one of the legislators who supported the previous bill.
Public First Action is more supportive of state laws and has fought efforts to preempt them, but Carson said that if Washington “can come up with a comprehensive federal approach to these issues, preempting is a natural part of the constitutional order.”
Republicans on Capitol Hill have tried several times to preempt the state law, but they have failed. House Majority Leader Steve Scalise (R-Louisiana) told CNBC that state laws “stifle innovation” and that overriding state laws “is fundamental to everything we do.”
“There’s definitely bipartisan opposition to pre-empting,” said Rep. Ted Lieu (D-Calif.), co-chair of the House Democratic AI Committee, but noted that many Democrats have recently supported the Online Safety for Children Act, which would set federal privacy standards as a floor.
