Campaigns and PACs spent over $137M promoting AI-made midterm ads, most with no label
Campaigns and PACs spent over $137 million on AI-generated or AI-enhanced midterm ads, Wesleyan and AdImpact findings show. Most carried no AI label, and some showed real officials in scenes that never happened, the New York Times reported.
The Watch · Security desk
Wesleyan's count of at least 164 unique ads found just under $80 million. AdImpact's count, limited to ads with an identifiable disclosure, found slightly more than $13 million. The broader-criteria total is more than $137 million.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Wesleyan count, at least 164 unique ads | $80M just under | 10 | |
| AdImpact, disclosed ads | $13M over | 11 | |
| Broader-criteria total | $137M over | 9 |
What happened
- A Bruce Blakeman campaign video placed New York Mayor Zohran Mamdani and Gov. Kathy Hochul together in a synthetic ad for a fictional "political delusion pill", with no AI disclaimer, the Times reported.
- A Make Oklahoma Great Again ad showed a synthetic Democratic gubernatorial candidate embracing Hillary Clinton, backed by more than $200,000 in Oklahoma television buys tracked by AdImpact.
- Republican campaigns and Republican-aligned PACs produced more than 80 percent of the reviewed ads, according to the Wesleyan Media Project's broader dataset.
- In a complaint to the state Board of Elections, New York's state Democratic Party alleges the Blakeman ads violate a law that requires AI labels on political communications that are materially deceptive.
Why it matters
- Who's exposed exposure Teams that protect public-facing leaders now have a documented example of a campaign video using a sitting mayor's and governor's voices and likenesses with no disclaimer.
- What it limits constraint A missing label is no evidence of authenticity. Most ads in the sample carried none, and roughly half of states require one, so labels cannot serve as a filter.
- What it sets up precedent A ruling on the New York complaint would show whether a satire claim exempts a fabricated ad from a state labeling law.
Next story · Invest · 26m ago
Nasdaq drops 2.15% as CoreWeave swaps imply over 60% odds of default in five yearsConfirmed2 publishers
The $137 million is the widest of three counts. A separate Wesleyan Media Project accounting, covering at least 164 unique ads, found just under $80 million in spending [10]. AdImpact's count, limited to ads with an identifiable disclosure, came to slightly more than $13 million [11]. Gadget Review notes that the $137 million uses broader inclusion criteria and that the three figures are not interchangeable [9].
The broad total includes routine production work. Campaigns used AI to animate still photographs, generate crowd scenes, recreate childhood settings and replace conventional stock footage [8]. In our view the $137 million is a ceiling on the fabricated material, not a count of it [1][8].
Other examples reviewed by the Times showed Abdul El-Sayed laughing at cards containing extremist messages and James Fishback at a gathering with Rep. Ilhan Omar and Iranian leaders [5]. Neither event occurred [5].
About 70 percent of ads in the broader review did not disclose AI use; in the narrower set, 31 percent did [7]. The Wesleyan researchers said the true total is likely higher than their counts, since ads in states with no disclosure requirement are harder to find [12].
Roughly half of U.S. states require disclosure, according to CNN, and the rules vary in scope [16]. Michigan requires a three-second voiceover acknowledging AI use alongside a four-second written disclosure [16]. A campaign that complies in one state may face a legal complaint in another [17].
The Blakeman campaign said the ad was satirical and required no disclaimer [15].
Gadget Review concludes that AI political advertising is no longer a marginal experiment [13]. The reporting does not break out how much of the total went to fabricated depictions of real people, and it does not address impersonation outside politics.
What to watch
- The New York Board of Elections response to the state Democratic Party's complaint over the Blakeman ads.
- Any revision of Wesleyan's counts, which its researchers say understate the total because ads in states without disclosure rules are harder to detect.
- A second state complaint that tests the gap between Michigan-style disclosure rules and states with none.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence62
- Adoption55
- Hype gap+25
- Incentives58
- Confidence64
Perspective Coverage
3 publishers- Builder
- Builder 17%
- Operator
- Operator 50%
- Investor
- Investor 33%
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Political campaigns and PACs spent more than $137 million promoting AI-generated or AI-enhanced advertisements during the 2026 midterm cycle, according to findings from the Wesleyan Media Project and AdImpact reported by the New York Times.
- [2]
Most of the AI-generated or AI-enhanced ads carried no disclosure that artificial intelligence was involved.
- [3]
A Bruce Blakeman campaign video fabricated New York Mayor Zohran Mamdani and Gov. Kathy Hochul appearing together in a synthetic advertisement for a fictional 'political delusion pill', using their voices and likenesses without any AI disclaimer.
- [4]
A Make Oklahoma Great Again advertisement showed a synthetic Democratic gubernatorial candidate embracing Hillary Clinton, and AdImpact tracked more than $200,000 in Oklahoma television buys behind that ad alone.
- [5]
Other reviewed examples showed fabricated versions of Abdul El-Sayed laughing at cards containing extremist messages and James Fishback appearing at a gathering with Rep. Ilhan Omar and Iranian leaders; neither event occurred.
- [6]
More than 80 percent of the reviewed ads came from Republican campaigns or Republican-aligned PACs, per the Wesleyan Media Project's broader dataset.
- [7]
Approximately 70 percent of ads in the broader review carried no AI label; only 31 percent disclosed AI use in the narrower set.
- [8]
Campaigns used AI for production tasks such as animating still photographs, generating crowd scenes, recreating childhood settings, and replacing conventional stock footage.
- [9]
The $137 million total applies broader inclusion criteria than the other two counts, and the spending figures reflect different methodologies and are not interchangeable.
- [10]
A separate Wesleyan accounting covering at least 164 unique ads identified just under $80 million in spending.
- [11]
AdImpact's more conservative count, limited to ads with an identifiable disclosure, found slightly more than $13 million.
- [12]
Wesleyan researchers said their counts likely understate the real total, because ads running in states without disclosure requirements are harder to detect.
- [13]
AI political advertising is no longer a marginal experiment.
- [14]
The New York State Democratic Party filed a complaint with the state Board of Elections over the Blakeman advertisements, alleging violation of a state law requiring materially deceptive political communications to carry an AI label.
- [15]
Blakeman's campaign said the ad was satirical and required no disclaimer.
- [16]
Roughly half of U.S. states require campaigns to disclose AI-generated or manipulated material in political advertising, though the rules vary significantly in scope; Michigan requires a three-second voiceover acknowledging AI use alongside a four-second written disclosure.
- [17]
The requirements differ enough across states that a campaign compliant in one state may face a legal complaint in another.
Sources
3 independent publishers whose own reporting we read for this story.
- gadgetreview.comAI-Generated Midterm Ads Hit $137M, Most Without a Label - Gadget Review
1 article · October 16, 2026
- gizmodo.com$137 Million Has Been Spent on AI-Generated Ads Ahead of Midterms. Only 27% Included a Disclaimer
1 article · October 16, 2026
- techstrong.aiThe ‘AI Election’ Arrives: Campaigns Spend $137 Million on Synthetic Media as Misinformation Concerns Rise - Techstrong.ai
1 article · October 16, 2026
Topics and entities
Topics
- Deepfakes
- Election AI disclosure laws
- AI in political advertising
- 2026 U.S. midterm elections
Entities
- AdImpact
- Wesleyan Media Project
- American Association of Political Consultants
- Anthropic