Written by Lisa Fazio, Vanderbilt University; [This article first appeared in The Conversation, republished with permission]
Zoran Mandani plays a creepy trick-or-treater, Gavin Newsom body-slams Donald Trump, and a sombrero-wearing Hakeem Jeffries. This is not an elaborate joke setup. Instead, these are all examples of recent political videos generated by AI. New, easy-to-use tools, and the acceptance of those tools by politicians, means these fake videos will quickly become commonplace in American politics.
Perhaps the most interesting thing about many of the videos is how obviously fake they are. Rather than trying to trick the viewer into believing that the events depicted actually happened, the video serves a different purpose. President Trump did not post a video of himself wearing a crown and riding a fighter jet dumping feces on protesters because he wanted people to believe that the flight actually happened. Perhaps he did it to express his feelings about the protests and create inside jokes with his supporters.
Concerns about the political implications of AI-generated videos have been widespread since the term deepfake was coined in 2017. Steady advances in technology mean that the distinction between real and fake can become a serious threat. However, the use of AI images today is mainly aimed at making money by creating memes, typical social media content.
Rise of talent
Internet platforms use algorithms designed to keep people engaged. This usually means promoting content that resonates emotionally. AI-generated political videos often elicit emotional responses such as amusement or anger.
People tend to share information that touches their emotions. For example, people are more likely to pass on urban legends that cause disgust, and emotionally moving news stories are more likely to be on the New York Times' list of most emailed articles. A similar pattern occurs online, where emotional content is more likely to be spread than non-emotional content.
Additionally, strong emotions can interfere with our ability to detect misinformation. People have a harder time distinguishing truth from falsehood in political news headlines when they are experiencing stronger emotions, such as enthusiasm, excitement, or fear. Therefore, an emotionally appealing AI-generated video is more likely to go viral, but may reduce people's ability to tell what's real and what's fake.
online politics
Creating and sharing AI videos is also a powerful way for people to demonstrate their allegiance and political identity. “I’m a Trump supporter, so I’ll post an AI video of ICE detainees screaming to own a library,” or “I’m a Democrat, so I’ll share an AI video of Governor Newsom with J.D. Vance talking about the couch to show I’m in on the joke.”
What's new in recent months is that not only advocates but also campaigns and politicians are using AI-generated videos. According to a New York Times analysis, President Trump often uses AI imagery to “attack his enemies and incite his supporters.”
These new tools also enable active participation in the political process. Rather than simply watching politicians and voting, citizens can take an active role in shaping the conversation between elections.
Information technology researcher Kate Starbird writes about a similar dynamic in how ordinary Americans found “evidence” of voter fraud in the 2020 election. Politicians told people that voter fraud would occur, and when voters saw things they didn't understand when voting, such as using a Sharpie pen to mark their ballots, they interpreted that behavior as evidence of voter fraud. Politicians then spread the evidence online to support the false story.
New AI tools make this cycle of participatory disinformation even simpler. Instead of reinterpreting actual events as evidence for false claims, people can easily generate that evidence themselves.
Loud AI video
AI video creation tools make it incredibly easy for people to create hundreds of videos in bulk, simply post them online and see which content becomes popular and goes viral. In fact, that's exactly what appears to have happened in an AI-generated video of a recent Immigration and Customs Enforcement raid. According to an investigation by 404 Media, Facebook user “USA Journey 897” posted various real-life videos of police operations, as well as absurd AI videos of people carrying whales and riding tigers.
However, after the release of a new version of OpenAI's Sora video generator on September 30, 2025, this account completely changed to posting multiple fake videos about deportation every day. Most videos have racked up hundreds of thousands of views, with one fake video of a Walmart employee being taken into custody more than 4 million times.
These accounts are typically hosted overseas and exist to earn money through creator incentive programs. These incentives create an environment in which social media no longer informs people about the world, but instead acts as an entertainment mirror, presenting us with the world we want to see, or at least a version of the world that captures our attention and anger.
AI-generated political ads are pushing ethical boundaries.
flow into the internet
It's not always easy to determine which videos are real and which are generated by AI. A recent audit by the publication Indicator found that platforms regularly fail to properly label AI content. The researchers posted more than 500 AI-generated images and videos on Instagram, LinkedIn, Pinterest, TikTok, and YouTube. Less than a third of posts were properly labeled as AI-generated, and even posts generated by the platform's own AI tools were often missed.
For years, a big concern about political deepfakes was that they tricked people into thinking something had happened when it didn't. That could still happen, but for now, AI-generated political videos are a mix of entertainment, memes, legitimate attempts at persuasion, and attention-getting ways to make money.
In other words, they are now just like the rest of the internet. Most of what we watch and share is meant to be entertaining, some is meant to inform and persuade, and much else exists solely to monetize our attention.
Lisa Fazio, Associate Professor of Psychology; Vanderbilt University
This article is republished from The Conversation under a Creative Commons license. Read the original article.
