AI Firms’ Election Spending: What They Want

The burgeoning artificial intelligence industry is making a significant push into the political arena, pouring millions into election campaigns to influence the development of AI legislation in Congress. As of the end of June, two prominent AI-focused Political Action Committees (PACs) have collectively invested at least $44 million in 40 House and Senate candidates. This early spending is a preview of the more than $200 million these groups aim to deploy through the remainder of the primary season and into the general election, according to their fundraising disclosures.

This aggressive political engagement positions the AI sector as an increasingly influential force in Washington, seeking to shape the contours of the nation’s first comprehensive AI regulatory framework. The urgency stems from the rapid advancement and deployment of powerful AI models, exemplified by systems like Claude Fable and Mythos, which have raised both excitement and concerns regarding their capabilities and potential risks.

Brad Carson, who leads Public First Action, a non-profit organization overseeing several PACs, has observed a marked increase in AI-related legislative proposals and discussions. “They offer significant benefits, but also substantial dangers, and it’s imperative to have government oversight,” Carson told CNBC. He emphasized a bipartisan consensus on this front, stating, “From the right to the left, pro-Trump to anti-Trump, everyone recognizes that.”

Josh Vlasto, co-leader of Leading the Future, underscored the critical importance of establishing the right regulatory architecture now. “It is absolutely vital that we address this immediately and with urgency, as the technology is still in its nascent stages but is being adopted at an unprecedented scale,” he stated.

The strategic investments by these AI PACs are already yielding results, with a high success rate in primary elections. Leading the Future has backed 28 candidates, 25 of whom have secured their party’s nomination. While two races are still pending, only one candidate, Jesse Jackson Jr., lost. The group also opposed Alex Bores in New York’s 12th Congressional District, who ultimately did not win the Democratic primary.

Public First Action has supported candidates in 11 races, with all but one, Alex Bores, emerging victorious from their primaries. Carson indicated that Public First Action intends to expand its engagement to 50-60 races by the end of the midterm elections.

This playbook mirrors that of other emerging industries. In the 2024 election cycle, the crypto-backed PAC Fairshake allocated a substantial $200 million, championing pro-crypto candidates. This investment contributed to the passage of a significant stablecoin bill and advanced key legislation for digital assets favored by major players like Coinbase and Ripple.

Leading the Future has so far expended over $24 million in primary races, as reported to the Federal Election Commission. By the close of 2025, the group had amassed $125 million in contributions from a diverse set of donors, including 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 disbursed $20 million to date and announced raising $80 million through June. While a spokesperson confirmed a $20 million donation from Anthropic, these funds are specifically designated for public education on AI policy and are not for direct political contributions. Public First Action does not publicly disclose its donors, though Carson noted contributions from employees at OpenAI, Google, DeepMind, and X.

**AI’s Dual Approach to Shaping Policy**

The two leading AI advocacy groups, Leading the Future and Public First Action, have encountered each other in several races, sometimes spending against opposing candidates and engaging in public debates. Despite these competitive instances, their policy differences are more nuanced than a simple pro- or anti-regulation stance. Both groups generally support implementing safeguards and agree on issues such as protecting children online.

A key divergence lies in the question of whether a single federal standard should supersede state-level AI regulations. Leading the Future advocates for a “broad, national, consistent framework for regulation governing AI,” according to Vlasto. He clarified that this stance does not inherently oppose state laws, citing the group’s support for New York’s RAISE Act, a landmark AI law co-authored by Alex Bores.

However, the complexities of the RAISE Act highlight the group’s intricate position. Leading the Future reportedly spent approximately $8 million opposing Bores, largely due to his push for a more stringent version of the RAISE Act than what was ultimately enacted. Governor Kathy Hochul’s intervention led to amendments that softened reporting requirements and penalty structures, aligning the New York law more closely with California’s. This resulted in Leading the Future endorsing the final legislation while simultaneously opposing a lawmaker who had initially supported its more assertive iteration.

Public First Action leans towards greater support for state-level regulations and has actively opposed efforts to preempt them. Carson, however, acknowledged that a “comprehensive federal approach” could naturally lead to preemption within the constitutional framework.

On Capitol Hill, Republicans have repeatedly attempted, without success, to establish federal preemption over state AI laws. House Majority Leader Steve Scalise has argued that state regulations “are hurting innovation” and that overriding them is “going to be the foundation of anything we do.” Representative Ted Lieu, a co-chair of the House Democratic AI commission, indicated that while there is “definitely bipartisan disapproval of preempting with nothing,” many Democrats recently supported a federal standard for online child safety privacy, suggesting a potential floor for federal regulation.