Silicon Valley's $125 Million Bet: How an AI Industry Super PAC Is Reshaping Republican Primaries

In the spring of 2026, voters in a half-dozen Republican primaries are receiving mailers and seeing television ads about immigration enforcement, healthcare, and support for Donald Trump. The ads contain no mention of artificial intelligence. But the money behind them comes almost entirely from the AI industry — and the candidates they promote share one common trait: opposition to state-level regulation of AI companies.

Leading the Future, a super PAC network launched in August 2025, has raised more than $125 million from AI industry executives and is deploying that money across congressional and gubernatorial races nationwide [1]. Its Republican-focused affiliate, American Mission, has already claimed wins in multiple Texas primaries and is expanding into Georgia, Montana, and Iowa [2]. A rival effort, Public First, backed by $20 million from the AI company Anthropic, is attempting to counter with candidates who favor stricter AI oversight [3].

The result is something without precedent in American politics: two competing factions of the same industry spending tens of millions of dollars to shape primary elections around a regulatory question most voters have never heard of.

The Money: Who Is Paying and How Much

Leading the Future's fundraising dwarfs most single-issue super PACs at comparable stages of a midterm cycle. FEC filings show the network raised $50.3 million through its main entity, with an additional $125 million across affiliated groups during the second half of 2025 [4]. It entered 2026 with $70 million in cash on hand — $39.3 million in the main PAC, $5.4 million in Think Big (its Democratic affiliate), and $5 million in American Mission (its Republican affiliate) [1].

Source: FEC Filings / CNBC

Data as of Jan 30, 2026CSV

The donor base is concentrated among a small number of AI industry figures. OpenAI President and co-founder Greg Brockman and his wife each contributed $12.5 million, totaling $25 million from a single household [4]. Marc Andreessen and Benjamin Horowitz, co-founders of the venture capital firm Andreessen Horowitz (a16z), each gave $12.5 million [4]. Other disclosed donors include 8VC founder Joe Lonsdale ($250,000 to American Mission), angel investor Ron Conway ($500,000 to Think Big), and the AI search company Perplexity [1].

On the opposing side, Anthropic's $20 million contribution to Public First Action represents the largest known single donation to a pro-regulation AI PAC [3]. Public First describes itself as "a counterforce that will defend the public interest against those who aim to buy their way out of sensible rule-making" [5].

For context, the crypto industry's super PAC Fairshake spent roughly $130 million during the 2024 cycle across both parties [6]. Leading the Future's $125 million haul in just six months suggests AI industry political spending could match or exceed that benchmark by November 2026.

The Races: Where the Money Is Going

Texas: The Testing Ground

Texas served as American Mission's first major proving ground. Groups funded by AI industry leaders spent more than $2.8 million in the state, mostly on U.S. House GOP candidates [7]. Leading the Future spent a combined $1.4 million supporting four Republican candidates — Jace Yarbrough, Jessica Steinmann, Chris Gober, and Tom Sell — all of whom won their primaries or advanced to runoffs [2].

Source: FEC Filings / NOTUS

Data as of Apr 1, 2026CSV

The largest single beneficiary was Chris Gober, a political lawyer who previously worked for Elon Musk's super PAC during the 2024 election, running to succeed retiring Rep. Michael McCaul. American Mission spent approximately $747,000 on his behalf through late February [7]. Jessica Steinmann received about $511,000 in ad spending for Texas' 8th Congressional District, while Jace Yarbrough received nearly $130,000 in North Texas' newly drawn 32nd District [7].

Expansion Into New States

Claiming momentum from Texas, Leading the Future announced a $1.5 million expansion into three additional GOP primaries [2]:

• Georgia's 1st Congressional District: Jim Kingston, an insurance executive and son of former Rep. Jack Kingston
• Montana's 1st Congressional District: Aaron Flint
• Iowa's 4th Congressional District: Chris McGowan

All three seats are safely Republican, meaning the primary winner is virtually guaranteed to join Congress. This makes the primary the decisive election — and outside spending more influential than it would be in competitive general election districts.

Florida: The First Gubernatorial Play

Leading the Future's most ambitious bet is a $5 million pledge to support Rep. Byron Donalds' campaign for Florida governor [8]. The race marks the PAC's first foray into a statewide contest. Donalds, endorsed by Trump and having already raised $50 million, is widely considered the GOP front-runner [8]. Florida Republicans are engaged in a protracted fight over AI legislation backed by Gov. Ron DeSantis but opposed by the industry — making the governor's mansion a strategic prize for companies seeking to block state-level regulation [9].

Other Targets

In North Carolina, the PAC spent six figures to help Laurie Buckhout advance in a Republican primary [2]. In New York, Leading the Future is spending to oppose Democratic state Assemblymember Alex Bores, an AI safety advocate running for the seat being vacated by retiring Rep. Jerry Nadler [1].

The Messaging: AI Ads That Never Mention AI

The most striking feature of these campaigns is what they don't say. As NBC News reported, ads funded by AI industry money are "about everything except AI" [10]. Polling consistently shows a majority of voters remain skeptical of AI and do not want data centers in their communities, so the PACs have adapted their messaging accordingly.

In the New York race against Bores, one ad stated: "It's black and white: Alex Bores' tech company works for ICE" — despite Bores having left his employer over ICE-related concerns [10]. In other races, ads focus on healthcare votes, support for Trump, or attacks on opponents' records on immigration.

Brad Carson, co-leader of the rival Public First, acknowledged the dynamic: "We know AI isn't the first thing on every voter's mind when they go to the polls" [10].

Co-strategist Zac Moffatt, a veteran Republican operative and CEO of Targeted Victory, stated that "our recent success across other primaries has allowed us to expand our footprint and continue supporting pro-innovation candidates who understand the need for a national regulatory framework on AI" [2]. Josh Vlasto, who previously worked for Sen. Chuck Schumer, leads the PAC's Democratic-side operations [11].

The Donors' Interests and Candidates' Positions

The policy objective uniting Leading the Future's donors is straightforward: preempt state-level AI regulation with a federal framework that the industry considers more favorable. The PAC has said it will support candidates who back "a national regulatory framework for AI, rather than allowing each state to set its own rules" [7].

This aligns directly with the financial interests of its donors. OpenAI, Andreessen Horowitz, and Palantir all face a patchwork of state-level proposals that would impose disclosure requirements, liability standards, and safety testing mandates on AI systems. A single federal framework — particularly one shaped by industry-friendly legislators — would reduce compliance costs and preempt stricter state laws.

The candidates receiving PAC support have broadly adopted these positions. Donalds, for example, has said that "data warehoused in Florida is safer" and has positioned himself as welcoming to the AI industry [9]. The Texas candidates backed by American Mission all favor lighter regulatory approaches [7].

Anthropic, meanwhile, supports a different regulatory model: preserving state authority until Congress passes stronger federal standards, along with model transparency requirements, export controls on AI chips, and targeted regulation of AI-enabled bioweapons and cyberattacks [3].

The Regulatory Vacuum

No federal statute specifically addresses AI in political campaigns [12]. The regulatory landscape is split between two agencies with different mandates:

The Federal Election Commission considered and rejected a rulemaking on AI in campaign ads in September 2024. Instead, the FEC adopted an Interpretive Rule clarifying that existing regulations on "fraudulent misrepresentation" apply regardless of the technology used [13]. This means deepfake videos of candidates making false statements are covered, but AI-generated ad copy, voter targeting algorithms, and persuasion optimization tools are not specifically regulated.

The Federal Communications Commission proposed separate rules requiring on-air disclosure when television and radio political ads contain AI-generated content, along with documentation in broadcasters' public political files [14]. These proposed rules apply only to broadcast media — not to digital ads, social media, or direct mail, where most modern campaign spending occurs.

At the state level, eleven states — including California, Texas, Michigan, and Washington — have enacted laws regulating AI-generated deepfakes in political advertising [12]. Similar legislation is pending in 28 additional states [12].

The Campaign Legal Center has argued that existing frameworks are insufficient: "AI tools allow political campaigns to micro-target voters based on psychological profiles derived from online behavior, often without adequate consent" [15]. The Brennan Center for Justice has separately warned about AI's potential for voter suppression through targeted disinformation [16].

Defenders of the current framework point to First Amendment concerns. The FCC has tentatively concluded that disclosure requirements pass constitutional scrutiny, but broader prohibitions on AI-generated political speech raise unresolved questions about government regulation of campaign communications [14].

The Counterfactual Question: Did AI Money Actually Decide These Races?

Leading the Future's claim of "early wins" in Texas primaries deserves scrutiny. The PAC spent $1.4 million across four candidates who all won or advanced — but isolating the effect of that spending from other factors is difficult.

Several of the Texas races involved open seats with crowded primary fields, where even modest outside spending can shift outcomes. Chris Gober, who received $747,000 from American Mission, was also a well-connected political operative with existing name recognition in Republican circles [7]. Jessica Steinmann's $511,000 in support came in a district where no candidate had a dominant fundraising advantage.

Political scientists have long debated the marginal effect of super PAC spending in primaries. Research on the crypto industry's 2024 Fairshake PAC found that its spending was most effective in low-information primaries where voters had few other cues to distinguish candidates [6]. The same dynamic applies here: in safely Republican districts where the primary is the real election, a six- or seven-figure ad buy from an outside group can be decisive simply because it creates an information asymmetry.

The question of whether AI-driven campaigning tools — as opposed to AI industry money — played a role is harder to answer. There is no public evidence that Leading the Future or American Mission used AI-powered voter targeting or ad generation tools that differ materially from the data analytics platforms used by conventional super PACs. The PAC's advantage appears to be financial, not technological.

The Democratization Argument

Supporters of AI-financed super PACs argue they lower barriers for non-establishment candidates. Jim Kingston in Georgia, Aaron Flint in Montana, and Chris McGowan in Iowa are not incumbents and lack traditional donor networks [2]. Outside spending from Leading the Future gives them financial competitiveness they would not otherwise have in crowded primaries.

Zac Moffatt has framed the PAC's work as supporting "pro-innovation candidates" who might not otherwise be viable [2]. This echoes arguments made by the crypto industry's Fairshake PAC in 2024, which successfully boosted several first-time candidates into Congress.

Critics counter that this is not democratization but substitution — replacing one set of donor dependencies with another. A candidate who wins a primary because of $500,000 in AI industry spending arrives in Congress owing a political debt to that industry, regardless of whether they had a traditional donor network before.

Carson of Public First put it more bluntly: "Polling consistently shows significant public concern about AI and overwhelming voter support for guardrails" [5]. If outside PAC money allows candidates opposed to those guardrails to win primaries where pro-regulation candidates might otherwise prevail, the argument goes, the PAC is overriding rather than reflecting voter preferences.

Second-Order Effects on the Republican Party

If AI-optimized PAC spending consistently outperforms traditional methods in primaries, it could reshape the GOP's candidate pipeline. The pattern emerging from early 2026 races suggests a specific profile: candidates who favor federal preemption of state AI regulation, are comfortable with light-touch industry governance, and align with the broader tech-friendly faction of the Republican coalition.

This profile is not ideologically uniform — Donalds is a Trump-endorsed conservative, while some Texas candidates come from more establishment backgrounds. But the common thread is opposition to the regulatory frameworks favored by figures like DeSantis, who has positioned himself as an AI skeptic [9].

If the pattern holds through November, the 2027 Congress could include a cohort of Republican members whose elections were substantially financed by AI companies — creating a built-in constituency against regulation in a chamber that will face mounting pressure to act on AI governance. Whether that constitutes industry capture or legitimate political organizing depends on one's view of the role super PACs play in American democracy.

What Remains Unknown

Several questions remain unanswered by available evidence. Leading the Future has not disclosed which specific AI tools or vendors, if any, it uses for voter targeting, ad creative, or persuasion optimization. FEC filings require disclosure of expenditures but not the algorithmic systems that inform spending decisions. The nonprofit arm of the network, Build American AI, operates with even less disclosure as a 501(c)(4) entity [1].

The precise number of voters who have received PAC-funded communications, and the demographic or psychographic profiles used to target them, is not publicly available. This information gap makes it impossible to assess whether the PAC's targeting strategies differ in kind from conventional political operations or represent something qualitatively new.

Academic research on AI-powered political persuasion suggests the technology has measurable effects. A 2024 study published in the Proceedings of the National Academy of Sciences found that personalized political ads tailored to individuals' psychological profiles were more effective than generic ads, and that such personalization could be "automatically generated and validated on a large scale" [17]. Whether Leading the Future or its affiliates use such methods is unknown.

The 2026 midterms are testing whether an industry can buy its way into a favorable regulatory environment through primary elections. The answer will shape not only AI policy but the broader question of how emerging industries interact with American democracy.