Two years ago, AI-generated political advertising barely registered as a line item. This cycle, it is a nine-figure industry flooding television screens and social feeds from Texas to New York.
According to new data from the Wesleyan Media Project, political campaigns and groups have spent more than $137 million airing ads created at least partly with AI tools ahead of the 2026 midterm elections. For comparison, the group identified less than $50,000 in such spending during the entire 2024 cycle.
"I think what we're seeing with A.I. has surprised even us," Erika Franklin Fowler, co-director of the Wesleyan Media Project, said in a statement reported by Gizmodo. "A.I. can be used in so many different ways that it creates new opportunities for advertising."
How Much Is Being Spent on AI Campaign Ads in 2026?
More than $137 million has been spent on AI-generated political ads ahead of the 2026 midterms, according to the Wesleyan Media Project. Researchers identified more than 220 videos using AI-generated fakery. Republican candidates and groups accounted for 81 percent of the ads and 88 percent of the spending.
The partisan gap is striking. As Gizmodo reported, Republicans are leading the spending while Democrats appear more wary of potential backlash. The New York Times documented one ad fabricating the voices and likenesses of Mayor Zohran Mamdani and Governor Kathy Hochul, released with no disclaimer that it was AI-made.
The Washington Post highlighted similar cases: a spot shared by Republican Ken Paxton in Texas's Senate race depicting Democrat James Talarico reading extreme views from a Bible, and an ad from Republican Bruce Blakeman in New York showing Hochul robbing taxpayers. Both were obvious fabrications. Neither depicted candidate was amused.
Are AI Political Ads Required to Have Disclaimers?
No comprehensive federal rule requires it. The Wesleyan Media Project found that just 27 percent of AI-generated ads this cycle included a disclaimer informing viewers of the synthetic content. Thirty-one states have banned or regulated AI deepfakes in campaign ads, creating a patchwork of rules across the country.
California offers one example of the state-level fight. As ABC10 reported, Republican gubernatorial candidate Steve Hilton's campaign has used AI-generated ads featuring Democrat Xavier Becerra, prompting questions about compliance with state disclosure laws.
Not everyone wants Washington to step in. Writing in the Washington Post, Max Raskin argued that concocting a new body of election law is "a waste of time at best and an encroachment on liberties at worst." With Congress divided over AI policy, as Newsweek's AI Policy Scorecard shows, federal action before November looks unlikely.
Why Are Platforms and Donors Pulling Back?
Meta quietly received $25 million back from political groups it had funded, shrinking its $65 million effort to elect AI-friendly state politicians by roughly 40 percent, according to filings reported by the New York Times. The company noted that more than 90 percent of its supported candidates won their primaries.
Meanwhile, Reddit is moving the other direction on civic integrity. MediaPost reported that the platform is relaunching its "Up The Vote" hub, partnering with National Voter Registration Day, and requiring political advertisers to participate in an AMA before running ads. Any ad containing AI-generated content on Reddit requires a label.
The stakes extend beyond domestic campaigns. The New York Times reported that researchers have identified waves of AI videos targeting Democrats in competitive Senate races linked to Russian influence operations, with activity expected to ramp up as Election Day approaches. A separate lawsuit filed by the League of Women Voters, covered by Bloomberg Law, alleges the Department of Homeland Security's voter data collection effort violates the federal Privacy Act of 1974.
What Does This Mean for Voter Outreach Strategy?
The AI ad boom confirms what smart operatives already know: voters are drowning in synthetic content, and authenticity is becoming the scarcest resource in politics. A survey released by the AAPC Foundation found that 95 percent of likely 2026 voters expressed concern about AI in campaigns, according to Campaigns and Elections.
That skepticism is an opening. When voters cannot trust what they see on a screen, a real conversation carries more weight than ever. Campaigns investing in genuine voter contact through tools like HyperPhonebank can cut through the noise with live, human outreach that no deepfake can replicate.
Data discipline matters too. With synthetic media muddying the information environment, campaigns need reliable intelligence on what voters actually think. Solutions like HyperPulse synthetic polling help strategists separate signal from slop without waiting weeks for traditional field work.
The Wesleyan numbers tell the story of a technology adopted faster than the rules, or the public, can keep up. The campaigns that win in November will be the ones that use AI wisely for efficiency while keeping their voter contact unmistakably human. To build that kind of operation, explore our campaign services or follow our ongoing campaign strategy coverage.