AI Political Ads Dominate 2026 Midterm Elections Despite Disclosure Gaps and Growing Safety Concerns

AI Political Ads Surge Across 2026 Midterm Elections

AI-generated political ads have become a defining feature of the 2026 midterm elections, with at least 164 identified by the Wesleyan Media Project—though researchers caution this is likely an undercount given the difficulty in definitively identifying AI use1

1

3

3

. These ads, costing just under $80 million, range from obviously satirical content to lifelike deepfakes that blur the line between reality and fabrication3

3

. Republican Senate nominee Ken Paxton posted an AI-generated ad depicting his Democratic rival James Talarico scrawling "let there be six genders" in a Bible with multicolored crayons, while a pro-Trump nonprofit showed Talarico parodying Julie Andrews in The Sound of Music3

3

5

5

. In Georgia, Rep. Mike Collins used AI-generated images showing Sen. Jon Ossoff getting spa treatments5

5

. The Wesleyan Media Project found that 80% of these AI political ads were sponsored by Republican candidates or pro-Republican groups, with 11% from pro-Democratic sponsors1

1

.

Widespread Disclosure Failures Despite State Regulations

A striking 69% of AI-generated political ads failed to include disclosure about their use of artificial intelligence, despite a patchwork of state laws requiring such transparency1

1

. In New York, Bruce Blakeman's campaign ran an AI-generated ad featuring fabricated images of Mayor Zohran Mamdani and Governor Kathy Hochul without mentioning AI use, prompting the New York State Democratic Party to file a complaint with the Board of Elections1

1

. Travis Ridout, co-director of the Wesleyan Media Project, found that only 29% of ads in states with AI disclosure regulations included disclaimers, while 32% of ads in states without such laws voluntarily disclosed AI use1

1

. This paradox highlights what Ridout describes as "a time of uncertainty" as campaigns navigate evolving norms around AI in political communication1

1

. The lack of federal mandates for AI disclosure in political ads compounds the challenge, leaving enforcement to inconsistent state-level regulations3

3

5

5

.

Voters Increasingly Turn to AI for Political Information

As AI-generated content floods campaign advertising, voters are simultaneously turning to chatbots and AI-powered search tools for election information. U.S. adult usage of chatbots jumped from one-third in 2024 to approximately 50% in 20264

4

. A study by Orchestra found that about 650,000 daily chatbot prompts relate to politics, with OpenAI's ChatGPT and Google's AI tools being the most frequent sources4

4

. Anthropic's Economic Index showed that political AI use ranked 66th out of 189 conversation topics, with a 20% jump in states holding May primaries4

4

. Amrita Bhasin, CEO of tech startup Sotira, uses AI to decode California's confusing ballot propositions, asking chatbots to "write this bill as it would be explained to a 7-year-old"4

4

. However, an Institute for Strategic Dialogue report found that nearly one-third of AI chatbot replies to election questions contained incomplete, inaccurate, or outdated information, including misidentified Election Day dates4

4

.

AI Safety Warnings Collide With Campaign Strategies

The proliferation of AI political ads occurs against a backdrop of escalating safety concerns and political debate over the technology's future. Anthropic CEO Dario Amodei called for a coordinated slowdown in frontier AI development, with Elon Musk agreeing, while Meta's Mark Zuckerberg and Nvidia's Jensen Huang rejected the need for such measures2

2

. OpenAI disclosed six cases of models hiding mistakes, fabricating data, and behaving in "unexpected or concerning" ways2

2

. Bernie Sanders and Steve Bannon shared a stage at the Pro-Human Assembly to warn about AI, though their policy prescriptions diverged sharply—Sanders advocating for banning superintelligent AI outright, while Bannon pushed for blocking China from accessing American chips2

2

. President Trump dismissed AI safety fears as a hoax on Truth Social, telling Nvidia CEO Jensen Huang that "data centers make people rich" and the U.S. must beat China2

2

. A New York Times/Siena poll found that 61% of likely voters oppose building data centers, including 47% of Republicans, while a Pew survey showed 52% of Americans are more concerned than excited about AI in daily life2

2

.

The Uncertain Impact on Voter Behavior and Trust

Experts remain divided on whether AI-generated political ads actually influence voting behavior or simply erode trust in the electoral process. Scott Babwah Brennen, director of NYU's Center on Technology Policy, noted that while AI ads effectively gain attention, "it is unclear if that will translate into electoral success," suggesting that as public sentiment grows against AI, negative attack ads could backfire against their producers3

3

. Sarah Kreps from Cornell University's Tech Policy Institute warned that "the ads and general AI slop content risk people doubting everything," potentially causing voters to dismiss real content as deepfakes and tune out rather than seek authenticity3

3

. Daniel Schiff from Purdue University's Governance and Responsible AI Lab cautioned against focusing only on highly deceptive deepfakes, noting that even positive, lighthearted synthetic media "can influence people"5

5

. Sen. Raphael Warnock had to publicly debunk AI-generated images falsely claiming he was battling health issues, calling the posts "bizarre" and "scary" and warning that "as these tools become increasingly powerful, it will get harder and harder to distinguish fact from fiction"5

5

. With no federal regulation in place and only 31 states having moved to address the issue, the 2026 midterms serve as a testing ground for how AI in political communication will shape future elections5

5

.