With election season 2026 rapidly approaching, it’s crucial to stay informed. From voting logistics to candidate stances, there’s a lot of information to take in — and if you’re a Gen Zer, it’s likely you’re getting this information from social media. But in a time when anybody’s name and likeness can be plugged into a generative AI platform to create fake photos, videos, and other forms of media and communication, it’s more important than ever to be critical of what you see online.
AI-generated political content has been both a tool and a weapon for political leaders and candidates over the last few election cycles. President Donald Trump is just one of many examples of political figures utilizing AI to their advantage (and to others’ detriment). Over the last few years, Trump’s team has produced fake content of multiple Democratic figures, such as House of Representatives Minority Leader Hakeem Jeffries, President Barack Obama, California Governor Gavin Newsom, and Senate Minority Leader Chuck Schumer, frequently presenting those he views as opponents in a negative light.
But it’s not just that AI is being used to create unflattering content about political figures — it can actually spread misinformation, which is dangerous any time, but especially during a crucial election. But unfortunately, political deepfakes are becoming increasingly easier to create thanks to how much personal data, photos, videos, and audio are available online from elected officials and candidates. “Deepfakes run on exposed data,” Harry Maugans, CEO of privacy protection platform Privacy Bee and cybersecurity expert, tells Her Campus. “Cloning a voice takes a few seconds of audio now, a convincing face model needs a handful of photos, and a candidate has hours of speeches, interviews, and podcasts sitting in public.”
Like it or not, AI misinformation now plays a crucial role in the current election cycle — and likely future ones as well. “Both parties have already run AI-generated ads this cycle, including a minute-plus deepfake of a Senate candidate where the ‘AI Generated’ disclosure was basically microscopic,” Maugans says. As AI technology continues to evolve, it’s becoming increasingly difficult to spot what content is real or fake. However, that doesn’t mean it’s impossible to spot. There are common mistakes that AI models make that are easy to detect if you know what to look for, as well as other signs that the content you’re seeing may need to be fact-checked.
Watch closely For Glitches & Mistakes
Even though AI technology continues to develop, it still isn’t flawless. It’s pretty well known that AI models frequently struggle to accurately depict hands, but Maugans points out other tells, including “warped … jewelry, teeth that blur together, hair melting into the background, lighting that doesn’t match the room, lip-sync running half a beat behind audio.” One of the biggest giveaways? “Watch what happens when a hand crosses the face or the head turns to profile, because face swaps tend to break right there,” Maugans says.
But even with all that information in mind, some AI models are advancing quickly. Maugans acknowledges that this is distressing. “The uncomfortable truth though is that the tells are disappearing; the models improve every few months,” he says. That means you can’t trust a video isn’t AI just because you can’t spot any mistakes. According to Maugans, you should “treat a glitch as a lucky break when you catch one,” but that shouldn’t be your only point of defense.
Gut-Check & Fact-Check Everything
In a time where anything online could be fabricated, it’s important to only trust reliable sources. “[Information on] voting logistics should never come from social [media],” Maugans says, referring to things such as polling locations, important deadlines, and rules for mail-in ballots. “Go straight to vote.gov or your state’s Secretary of State site.”
And if you do see political information being shared on social media, you should always fact-check it against original sources. If you see a video on social media of a candidate giving a speech, for example, you should try to find the original, unedited video from sources such as C-SPAN. Similarly, if you’re seeing news being shared on social media, you should verify it with official news reporting and confirm it’s both accurate and current. “For coverage, I lean on AP and Reuters, and FactCheck.org and PolitiFact are both solid,” Maugans says. “Before sharing something, trace it back to the original and check the date — old content gets recycled constantly.”
The bottom line? Don’t take everything you see at face value. “Even 30 seconds of checking puts you ahead of most of the internet,” Maugans says.
Maugans also suggests trusting your instincts. “Gut-check your emotions,” he says. “If a post makes you furious in under five seconds, that’s not an accident — that’s the design!” Negative engagement is still good engagement for those creating and spreading AI misinformation, so if something comes across your feed that immediately elicits heightened emotions, it’s probably a good idea to check where it’s coming from.
Don’t Become Distrustful Of *All* Information
There are serious risks that come with AI-generated misinformation, especially when it comes to something as important as politics — and it’s vital to understand them in order to combat them. “The sneakiest damage is what researchers call the ‘liar’s dividend,’” Maugans says. “Once voters know fakes exist, real footage becomes deniable too. A candidate caught on tape [doing or saying something they shouldn’t] can just shrug and call it AI.”
And it’s not just about damaging candidates’ credibility — AI fakes are contributing to the culture of distrust that already exists in the country. “The damage I worry about most is corrosion. Nobody needs to believe any single fake. If enough voters decide they can’t tell real from fake anymore, they stop trusting all of it, and that broken trust is exactly what bad actors are after,” Maugans says.
Despite this, not all hope is lost — especially for Gen Z voters. Maugans believes that younger voters have an advantage over older generations when it comes to detecting AI-generated content. “[Younger voters] smell synthetic content better than their parents do,” he says. “The trap is overconfidence.” So, stay vigilant, stay curious, and then, as Maugans says, “Verify [information] on official sources, then go vote.”