The Liar's Dividend: Trump, AI, and the Death of Shared Reality
When everything could plausibly be fake, liars benefit — even when they're lying about footage from the National Archives.
Last week, President Donald Trump terminated trade negotiations with Canada over what he called a fraudulent advertisement. “They cheated on a commercial,” Trump said. “Ronald Reagan loved tariffs, and they said he didn’t. And I guess it was AI or something. They cheated badly. Canada got caught cheating on a commercial. Can you believe it?”
The ad in question, funded by the government of Ontario and aired during Major League Baseball’s World Series, features a voiceover from former President Ronald Reagan warning Americans about the dangers of tariffs. “High tariffs inevitably lead to retaliation by foreign countries and the triggering of fierce trade wars,” Reagan says in the ad. “Then the worst happens. Markets shrink and collapse. Businesses and industry shut down, and millions of people lose their jobs.”
It’s a powerful message, delivered in Reagan’s own voice. And it is not AI-generated, despite what Trump claimed. The audio comes from Reagan’s April 25, 1987 radio address to the nation on free and fair trade, delivered from Camp David. The full speech is available through the Reagan Presidential Library and on YouTube. You can watch it right now. The National Archives has the transcript. This is documented history, as real and verifiable as any presidential address from the last half-century.
Trump didn’t just dispute the ad’s framing or argue about context. He claimed Canada fabricated it. “I guess it was AI or something,” he said, before calling it “FRAUD” in all-caps on Truth Social and insisting that Reagan actually “LOVED TARIFFS FOR OUR COUNTRY.”
None of this is true. And that might seem like the end of the story — president lies, gets fact-checked, moving on. But something different is happening here, something researchers have been warning about for years. Trump isn’t making a mistake. He’s executing a strategy that becomes more effective every day that AI-generated content floods the internet. And he’s told us exactly what he’s doing.
The strategy
In September, Trump was asked about viral footage showing someone tossing something out an upper-story White House window. His press team had already confirmed to reporters that the video was real. Trump’s response? “No, that’s probably AI.”
Then he said something remarkable. “If something happens that’s really bad,” Trump told reporters, “maybe I’ll have to just blame AI.”
He wasn’t joking. He wasn’t speculating about a hypothetical future scenario. He was describing his actual strategy, out loud, to a room full of journalists. And he’s been doing this for a while now.
Back in 2023, the Lincoln Project released an ad called “Feeble” that showed Trump stumbling over words. The ad included footage of Trump looking unsteady, paired with a narrator questioning his strength. Trump’s response was to claim on Truth Social that “the perverts and losers at the failed and once-disbanded Lincoln Project, and others, are using A.I. (Artificial Intelligence) in their Fake television commercials in order to make me look as bad and pathetic as Crooked Joe Biden.”
The Lincoln Project told the AP that no AI was used in the spot. It was just regular video of Trump, edited the way political ads have been edited since the invention of videotape.
This is the pattern. Real video becomes “probably AI.” Documented footage becomes “fake.” Historical records become “fraud.” And each time Trump does this, he’s not trying to convince everyone. He just needs to convince enough people that reality is unknowable.
It’s only going to get worse
Here’s where the timing gets important. Trump is calling real Reagan footage “AI or something” at the exact moment when actual AI-generated Reagan videos are everywhere.
On September 30, OpenAI released Sora 2, an app that lets users create realistic videos from text prompts. Within days, it hit #1 on Apple’s App Store. And people immediately started using it to generate videos of historical figures — Martin Luther King Jr., Michael Jackson, Stephen Hawking, and yes, Ronald Reagan.
These aren’t the clunky, obviously fake AI videos from a few years ago. Solomon Messing, a professor at NYU’s Center for Social Media and Politics, put it plainly in an interview with NPR: “You can create insanely real looking videos, with your friends saying things that they would never say. I think we might be in the era where seeing is not believing.”
The Sora app’s explore page has featured AI-generated clips of Ronald McDonald fleeing police in a hamburger-shaped car, security footage of tech CEOs shoplifting, and Jesus cracking jokes about “last supper vibes.” Some of the content is meant to be funny. Some of it is deeply disturbing. Malcolm X’s daughter, Ilyasah Shabazz, was horrified to see AI-generated videos of her father “making crude jokes, wrestling with the Rev. Martin Luther King Jr. and talking about defecating on himself.” The Martin Luther King Jr. estate had to intervene after users created what they called “disrespectful depictions” of the civil rights leader.
So when Trump waves his hand and says the Ontario ad is “probably AI,” he’s landing that lie in a world where people have legitimate reasons to be confused about what’s real. Researchers have a name for this dynamic. They call it “the liar’s dividend.”
The liar’s dividend
Back in 2018, two legal scholars — Danielle Citron and Robert Chesney — published research warning about what they called “the liar’s dividend.” The concept is simple but chilling: as AI-generated content becomes more sophisticated, liars benefit even when they’re not using AI themselves.
“If the public loses faith in what they hear and see and truth becomes a matter of opinion, then power flows to those whose opinions are most prominent — empowering authorities along the way,” they wrote in the California Law Review. “A skeptical public will be primed to doubt the authenticity of real audio and video evidence.”
This is the key insight. The liar’s dividend doesn’t require actually creating deepfakes. It just requires a public that’s uncertain enough about what’s real that they’ll accept “it’s fake” as a reasonable explanation for anything inconvenient.
Hany Farid, a professor at UC Berkeley who studies digital forensics, explained it this way in an interview with the AP: “I’ve always contended that the larger issue is that when you enter this world where anything can be fake, then nothing has to be real. You get to deny any reality because all you have to say is, ‘It’s a deepfake.’”
With apps like Sora churning out thousands of realistic fake videos every day, Trump can point at a 38-year-old speech from the National Archives and say “probably AI,” and it doesn’t sound completely insane to people who’ve spent the last month watching AI-generated videos of dead celebrities flood their social media feeds.
A ridiculous lie
So what did Reagan actually say in that 1987 speech? And why is Trump so desperate to claim it’s fake?
The speech came at a complicated moment. Reagan had just imposed 100% tariffs on Japanese laptops, power tools, and television sets in response to what he said was Japan’s failure to enforce a semiconductor trade agreement. It was the first major trade retaliation against Japan since World War II. The Reagan Foundation and Trump both point to this context, arguing that the Ontario ad misleadingly uses clips from a speech where Reagan was announcing tariffs.
But if you read the full speech, Reagan makes his position crystal clear. He opens by saying that “imposing such tariffs or trade barriers and restrictions of any kind are steps that I am loath to take.” He calls the Japan situation “a special case” and says he wants to “lift these trade restrictions as soon as evidence permits.”
Then he spends the bulk of the address explaining why tariffs are bad policy. He talks about how the Smoot-Hawley tariffs of 1930 “greatly deepened the depression and prevented economic recovery.” He warns that when countries impose high tariffs, “homegrown industries start relying on government protection” and “stop competing.” He says that “high tariffs inevitably lead to retaliation by foreign countries and the triggering of fierce trade wars,” which causes markets to “shrink and collapse.”
Reagan explicitly calls out politicians who “want to go for the quick political advantage, who will risk America’s prosperity for the sake of a short-term appeal to some special interest group.” He says protectionism “destroys prosperity.”
PolitiFact looked at whether the Ontario ad misrepresents Reagan’s views and rated the Reagan Foundation’s complaint “Mostly False.” The fact-checkers found that while the ad does edit sentences together from different parts of the speech, Reagan’s overall message was unambiguously pro-free trade and anti-tariff.
Trump, meanwhile, claimed on Truth Social that Reagan “LOVED TARIFFS FOR OUR COUNTRY, AND ITS NATIONAL SECURITY.” This is the opposite of what Reagan believed and said, repeatedly, throughout his presidency.
For those who would like to financially support The Present Age on a non-Substack platform, please be aware that I also publish these pieces to my Patreon.
So now what?
The standard advice for navigating this mess is to teach people how to spot AI-generated content. Look for weird hand movements, check for unnatural blinking patterns, watch for inconsistencies in lighting or shadows. Various organizations have put out guides and checklists. There are tools that claim to detect AI-generated images and videos.
This is all fine, I guess. But it misses the point entirely.
The problem isn’t so much that people can’t tell the difference between real and fake videos. The problem is that people don’t want to tell the difference when the fake confirms what they already believe, or when calling something fake protects someone they support.
Trump supporters aren’t going to watch the Ontario ad, pull up the Reagan Library’s YouTube channel, cross-reference the transcript from the National Archives, and conclude that their guy lied. They’re going to hear Trump say “it’s probably AI” and think, yeah, that sounds about right. Meanwhile, people who already distrust Trump aren’t the ones who need convincing.
This is how motivated reasoning works. We extend enormous benefit of the doubt to people who share our politics and apply maximum skepticism to their opponents. An AI detection guide doesn’t fix that. You can’t fact-check your way out of a problem that isn’t really about facts.
And here’s the thing — the technology is only going to get better. Whatever tells or artifacts exist in today’s AI-generated videos will be gone in six months. The detection tools will improve, sure, but they’ll always be playing catch-up. More importantly, they require people to actually use them, to actually care whether something is real.
What Trump has figured out is that in a world where everything could plausibly be fake, you can just declare things fake and move on. No investigation needed. No careful analysis of metadata or frame-by-frame examination. Just “it’s AI” and you’re done.
Toby Walsh, a professor of AI at the University of New South Wales, warned that this approach, telling the AP that this “leads to a dark future where we no longer hold politicians (or anyone else) accountable.” And that’s exactly the point. Accountability requires a shared reality. When powerful people can simply deny that reality exists, accountability becomes impossible.
I'm gonna push back on this "Death of shared reality" thing because in the world I live in (and in the world you live in, too) reality still means quite a lot. You and I still have to follow through on our promises, still have to deliver the goods, or there are consequences. And no amount of AI trickery lets us get off the hook for that.
The problem is confined to our politics, and even more specifically to the Republican Party. Yes, in that bubble-world, reality and truth means nothing. That's usually the case with people who have signed on to a cult. But please don't tell me that these lunatics have the ability to decide whether reality matters for the rest of us.
What artificial intelligence has done -- and I have no doubt this is the fulfillment of its clandestine purpose -- is give the ruling class the ability to rule in accordance with the slogan "ignorance is strength," which always seemed to me to be the darkest prophecy in the stygian darkness of George Orwell's ever-more-obviously prophetic "1984." What this means -- what is its truly bottomless horror -- is the Christonazi cabal that has conquered the former United States can now pursue its intended New Holocaust without fear of disclosure. In other words, the mere existence of AI enables Trump and his successors to forcefully dismiss as "just an AI fake" any documentation of their atrocities. The corrective function of visual journalism is thereby nullified forever.
As I said on Heather Cox Richardson’s 26 October (print) thread, technology is the "how" of the apocalyptic patriarchal dynamic I've been observing for years -- first noted in "Glimpses of a Pale Dancer," the foolishly optimistic, implicitly revolutionary, forward-looking work of photography and extensively footnoted text that was destroyed by arson on 1 September 1983, just as it seemed on the brink of major publication -- after which I never dared write of it again until I became a blogger.
Under patriarchy, any notion that history arcs toward justice (here used as synonymous with humanitarian socioeconomic and political liberty) is by far the most paralytically malignant Big Lie ever uttered. There is indeed an arc of patriarchal history, but any honest reading of its truths proves that -- aided and abetted by ever-more-efficient technologies of oppression -- it curves toward fulfilling the ultimate patriarchal purpose of ever-more-tyrannical omnipotence for our Masters and ever-more-inescapable slavery for all the rest of us. Hence my oft-repeated observation that -- both in purpose and consequence -- patriarchy is the cosmic version of smallpox-contaminated blankets.
(With heartfelt thanks to Ms. Molloy for what is probably the most important essay published on Substack thus far.)