Blakeman is releasing generative AI campaign videos, embracing trend experts say has little effect

In one AI-generated ad for gubernatorial candidate Bruce Blakeman, a made-up drug dealer, an undocumented immigrant and an energy CEO profess their support for his opponent, Democratic incumbent Gov. Kathy Hochul.

In another AI-generated ad, a fake Hochul says her administration raised taxes for “programs that matter, like free healthcare for sex workers.”

Blakeman, the Republican Nassau County executive, has joined a nationwide wave of politicians, including President Donald Trump, who are embracing often-outlandish AI-generated campaign ads. Throughout his campaign, Blakeman has used AI-generated content to depict Hochul and New York City Mayor Zohran Mamdani as "South Park" characters, thieves, professional movers and basketball players. Newsday identified at least 17 videos that appear to contain AI-generated content posted between Jan. 1 and July 17.

Blakeman describes the videos as a form of speech-protected satire.

“New Yorkers need a little humor to cope with Kathy Hochul’s endless tax and utility rate hikes, so the Blakeman campaign team has created these videos in partnership with our rapidly growing digital network,” Blakeman campaign spokesperson Madison Spanodemos said.

But experts say the low-cost and widely available tool isn't any more effective than typical negative campaigning with real people. Experts also say candidates risk inundating voters with “AI slop” — low-quality imagery easily identified as fabricated — that has become prolific on social media.

AI's intent

Fooling the electorate is not the intent with AI-generated ads, according to Yamil Velez, a Columbia University political science professor. 

“My sense is that what matters the most for an ad is whether it actually resonates and maybe provokes an emotional reaction, and AI can do that, but so can other forms of content, like non-AI-generated content," Velez said.

Gov. Hochul’s campaign has not used AI-generated figures in ads. 

"While Blakeman’s resorted to straight up lying to New Yorkers, Team Hochul doesn't need AI ads to tell them about him jacking up property taxes on Long Islanders more than once, letting violent crime hit decade highs on his watch, or running an armed MAGA militia that undercuts local cops – he did all that in real life," said Hochul campaign spokesperson Ryan Radulovacki.

New York does not have any laws banning AI use in campaign ads, though it does have a disclosure requirement.

Trump, a strong Blakeman ally, has begun using generative AI in social media posts, including a picture depicting himself as a Jesus-like figure, which was later deleted. The technology has been adopted by campaigns across the ideological spectrum, including Andrew M. Cuomo in the 2025 New York City mayoral election and the National Republican Senatorial Committee in ads attacking Texas Democratic primary candidates.

The ads, especially hyperrealistic ones, enable politicians to push boundaries on how they distort the truth to voters, according to Josh Tucker, the co-director of New York University’s Center for Social Media, AI, and Politics. 

“The idea in a democracy is that voters are presented with factually correct information about candidates, and then they make their choices," Tucker said. "And any time you have people who are trying to obscure those factually correct positions of candidates or actually correct things about candidates, that is, in a sense, undermining some of the normative basis of democracy.”

Tucker added that there is always a risk of voters believing deceptive campaign material, whether AI-generated or not.  

Though an all-out ban on generative AI in political communication is unlikely to hold up in court, "a more workable approach is clear disclosure and penalties when campaigns intentionally distribute synthetic material without flagging that it's AI-generated content," according to Velez.

Peter Loge, director of The George Washington University’s Project on Ethics in Political Communication, said voters must also punish candidates using AI to deceive by not electing them.

AI-generated ads may be more attractive to trailing candidates like Blakeman, according to Loge. These candidates might be willing to take bigger risks and likely have less money to spend on high-quality human ads. 

“My guess is candidates who feel like they’re losing will be more likely to try new techniques because they’ve got nothing to lose,” Loge told Newsday.

Velez, who researches how political campaigns are utilizing AI technologies, said campaigns take a risk when using these ads, especially as social media becomes increasingly oversaturated with AI-generated content.

“People are tired of AI slop, and I think that’s going to then start affecting these campaigns,” Velez said.

The uptick in AI-generated campaign ads comes two years after New York adopted a 2024 election law provision that requires disclosures in political communication using “materially deceptive media,” with an exception for parody and satire. New York's Division of Election Law Enforcement oversees compliance and did not respond to questions about whether any violations of this provision have been investigated. 

Many of Blakeman’s ads with AI-generated people have a disclosure at the end of the video, though not all of them do, including a June 13 video depicting Blakeman and Hochul as basketball players at a Knicks game.

The campaign did not answer questions about its AI disclosure policy or why the videos have discrepancies in disclosures.

Is it protected speech?

Some other states, including California, Texas and Michigan, have also adopted AI disclosure provisions in election law. 

"The consensus among First Amendment scholars is that speech produced with AI is still speech for purposes of the First Amendment," said Thomas Healy, a Seton Hall Law School professor.

Healy said these disclosure laws will likely be challenged in the future and may be struck down. The courts may be willing to allow AI disclosures given its approval of campaign finance disclosure laws, according to Francesca Procaccini, a Vanderbilt Law School professor.

The New York disclosure law has not yet been tested in court. 

Velez said whether these ads have mandated disclosures may not make much of a difference.

“My sense is those disclosures, they don’t have much of an effect,” Velez said. “But that’s also because I don’t really think the reason why these ads are effective is because they are pulling wool over people’s eyes. It’s because they’re trying to elicit a reaction.”

Andrew M. Cuomo’s 2025 mayoral campaign experimented with AI-generated performers in ads depicting Cuomo as a subway conductor, trader and stagehand. Cuomo also received criticism for an ad with an AI-generated Mamdani, which critics said played on racist and Islamophobic stereotypes. Cuomo’s campaign quickly deleted the ad and later said it was accidentally posted. Cuomo went on to lose the campaign.

Recent campaigns have used AI-generated portrayals of Gov. Gavin Newsom (D-Calif.), Gov. Gretchen Whitmer (D-Mich.) and Rep. Thomas Massie (R-Ky.) in attack ads. The Texas Democratic Senate primary also featured AI-generated content, with campaigns leveraging AI depictions of James Talarico and Rep. Jasmine Crockett.

Loge said AI-generated ads are just the latest in a long history of negative campaigning strategies.

“Candidates have always made stuff up about themselves and their opponents,” Loge said. “AI just makes it easier to do it at scale.”

 The issue with AI-generated ads is that they normalize politicians lying, which threatens democracy, according to Loge.

“[Voters] may just look at stuff and go, ‘Oh, I disagree with that,’ or ‘That looks sloppy,’ or ‘I don’t like it. It must be AI-generated,’ rather than looking at it going, ‘Oh, this is interesting. I wonder if that’s accurate. I wonder if my candidate is really doing a bad thing, or if the other candidate is really doing a good thing,’” Loge said.

He added, “If everybody thinks everybody’s lying all the time, then even when something is true, it’s easy to dismiss as a lie.” 

While campaigns adopting new technologies and capitalizing on social media have historically seen some success, Tucker said the evidence isn't there yet with AI-generated ads.

“I’m not sure we have the sort of quintessential success story yet in terms of a U.S. political campaign of someone who came out of nowhere and got a name for themselves through AI,” Tucker said.