Somewhere out there, a guy filmed himself "winning" $100,000 on a prediction market bet about Donald Trump saying "McDonald's" on TV. Trump never said it. The trade never happened. The website wasn't even real. According to a Wall Street Journal investigation, this was not a one-off embarrassment — it was allegedly the business model.
The Scam in Plain Numbers
The Wall Street Journal reviewed 1,105 videos from 10 creators posted between December 2025 and mid-May 2026, all promoting Polymarket, the prediction market platform. Roughly 70% of those videos showed bets that were not real. Across 118 of those videos, creators promoted nearly $900,000 in winnings that never existed — on positions that would have actually lost more than $166,000 if placed on live markets.
Read that again. Not "slightly misleading" or "lacking context." The trades did not happen. The money did not exist. The wins were fabricated and posted to social media for an audience that had no idea they were watching a performance with a fake set.
This is the fintech version of posing with a rented Lamborghini, except somehow more cynical, because at least the Lambo is a real car.
Clone Websites and Paid Silence
According to the Journal, the alleged mechanics here are genuinely brazen. Polymarket and its contractors reportedly built near-identical clone websites — including one called poiymarket.com, with an "i" swapped in for the "l" — where creators could simulate trades without putting a single dollar at risk. A fake platform, for fake trades, promoted as real wins to real people.
At least 10 creators allegedly received between $2,000 and $3,000 per month for this arrangement. The Journal reports they were instructed not to disclose the paid partnership. Some creators only added "@polymarket partner" to their profiles after the Journal started making inquiries. That is the social media equivalent of a restaurant putting up its health inspection score only after the health inspector walks in.
Federal law requires influencers to clearly disclose paid promotional relationships. The FTC has been inconsistent about enforcement, but the rule is not ambiguous. You cannot take money to promote a product and hide it from your audience. This is not a gray area.
The McDonald's That Wasn't
One creator, George Makihara, allegedly posted a January video implying he had won $100,000 after Donald Trump said "McDonald's" on television. The Journal reports that Trump did not say that in January. The video used older footage. The win did not happen.
This is almost impressively specific in its fabrication. It is not a vague "I made money on the market" claim. It names a president, a fast food chain, a specific on-camera moment, and a specific dollar figure. Every element was reportedly false. The audacity required to construct that video and post it publicly is, in a deeply grim way, kind of astonishing.
This Was Already a Platform With Problems
The influencer campaign reportedly landed on top of a platform already carrying significant credibility baggage. A Columbia University study estimated that roughly 25% of Polymarket's historical trading volume was likely wash trading — artificially inflated activity designed to manufacture the appearance of demand and liquidity that wasn't actually there.
A prior Wall Street Journal analysis found that 67% of profits on the platform went to just 0.1% of accounts. So the influencer videos were allegedly selling the idea of widespread, accessible winnings on a platform where, by the Journal's own earlier reporting, profits were already concentrated among a tiny sliver of users. The "I turned $500 into $100K" pitch was aimed at regular people on a platform where regular people were, statistically, losing.
Wash trading. Fake influencer wins. Concentrated profits. Clone websites. This is not a platform with a few rough edges. This is a platform where the alleged fraud appears to run in layers.
Why This Matters Beyond Crypto Twitter
Prediction markets had a genuine moment during the 2024 election cycle. Mainstream outlets quoted Polymarket odds like polling data. Serious people wrote serious articles about whether these markets were better forecasting tools than traditional surveys. The platform positioned itself as a truth-telling mechanism, a place where money on the line meant the crowd's real beliefs were surfacing.
If the Journal's reporting holds up, what actually surfaced was an allegedly coordinated campaign to make the platform look more active, more lucrative, and more accessible than it was. The people treating Polymarket odds as a signal were working with data generated, at least in part, by a system that allegedly had a significant fake-trade problem baked into its promotional strategy.
This is not an abstract concern about crypto regulation. Real people watched those videos and made real decisions based on them.
The Dingo Take
Look, prediction markets are not inherently evil. The underlying idea — that people with money on the line will aggregate information more honestly than pundits with nothing at stake — is intellectually interesting and has some empirical support. But that entire premise collapses the moment you introduce a coordinated campaign of allegedly fake trades, clone websites, and paid creators told to keep quiet about the arrangement. You cannot sell "the market knows the truth" while allegedly rigging the market's own promotional operation.
The thing that stands out here is not even the scale of the alleged fraud, though 70% fake trades across 1,100 videos is a staggering number. It is the specificity of the construction. Someone built a fake website with a nearly identical domain name. Someone wrote a brief for creators that included instructions not to disclose payment. Someone reviewed videos of a man claiming Trump said "McDonald's" and thought, yes, post that. This was not a few bad actors going rogue. According to the Journal, this was infrastructure.
The FTC should be looking at this. Hard. Because if the alleged mechanics described in the Journal's reporting are accurate, this is not influencer marketing gone a little sideways. This is an operation that allegedly built the tools, paid the people, and instructed them to lie to their audiences about money. That is not a content moderation problem. That is a fraud problem. And the people who watched those videos and opened accounts deserve better than a belated "@polymarket partner" in someone's bio.
