Fake or real, the “inside traders” on Polymarket are great engagement bait

In mid-March, conspiracy theories swirled claiming Benjamin Netanyahu had been replaced by an AI clone. Though there was no actual proof that the Israeli Prime Minister had been injured or killed, on X this spurred a flurry of posts promoting prediction markets where people bet on whether he would be out of office by March 31st. One newly created Polymarket account in particular caught the attention of bettors: dududududu22, which had purchased more than $177,000 worth of “Yes” shares at 4.7 cents. Surely, only someone with inside knowledge would take such a risky position?

“This makes him possible to get paid of $3,779,000 in case of win,” a post with a link to the Polymarket profile reads. Notably, the post is marked as a paid partnership.

On the market page, other users shout out dududududu22’s huge bet: “dudu please tell us something 😁,” reads one comment. “I want to buy because of dududududu22,” someone with the username “elonmusk911” commented. Dududududu22’s positions are currently worth just $1,889.53, after the price of “Yes” shares tanked to less than 1 cent. Because of Polymarket’s crypto foundations, the actions of dududududu22 are at once transparent and also keep them completely anonymous. Did they actually have insider knowledge of some kind, or were they going off of a hunch?

Insider trading on prediction markets has become one of the biggest storylines as platforms like Polymarket and rival Kalshi have exploded into the mainstream, generating hundreds of millions of dollars in trades on events like March Madness and geopolitics. Whereas insider trading is illegal for the stock market, for prediction markets, it’s sometimes touted as a good thing: suspicious bets (and massive wins) have been tied to the US kidnapping of Venezuelan president Nicolás Maduro and airstrikes on Iran. In February, Kalshi revealed it had taken action against an editor working for YouTuber MrBeast who had traded on related markets; Israeli officials recently arrested and charged several people including an Air Force major, who are accused of trading on Polymarket with insider information.

But claims of finding “insiders” are not always as they seem. The peer-to-peer betting nature of prediction markets means odds are ever-changing as other users buy and sell their positions. (The price of “Yes” and the price of “No” on a given event market fluctuate, but equal $1. “Yes” and “No” each being worth 50 cents means users believe the outcome is a toss-up; when the event happens, your shares are worth either $1 a piece, or nothing.) Polymarket and Kalshi push the idea that the “wisdom of the crowd” is more powerful than traditional sources of information. The crowd, as it turns out, can also be easily influenced — including by content paid for by prediction markets.

Despite bans on insider trading, content about it is great for juicing engagement. (Polymarket trading data is public, giving creators endless sources of things to post.) At every hour of the day, bettors on X flood the platform with attention grabbing posts that — depending on the person — are either more evidence of widespread, unmitigated cheating, or a hint at what to bet on next.

The posts are largely about Polymarket, the world’s biggest prediction market where users can bet on future events, like the outcome of sporting events or whether there will be layoffs in the tech industry (though it’s usually sports). An increasingly vocal opposition argues that what companies like Polymarket offer is really just gambling under a different name. But bettors are tapped in: On X, post after post claims to spot unusual activity suggestive of insider trading.

“BREAKING: A suspected military insider won $90k correctly predicting 9 separate military events! This guy is now betting big on US forces entering Iran!” one post reads. Follow-up posts about the account publicize the next bet, that US forces will enter Iran by March 31st.

“This guy is either very good at [open-source intelligence], very lucky, or might get info from people,” a tweet that includes the Polymarket profile says. It is also marked as a paid partnership.

There is often a juvenile, meme-like energy in these spaces, where finance guys and WAGMI crypto culture collide (“We’re All Gonna Make It” is technically speaking impossible with the basic function of prediction markets — one person winning necessitates that the other side loses). If TikTok influencers sell you a beautiful, aspirational lifestyle, prediction market influencers are selling you something more ruthless: the dream of monetizing life itself, of profiting when something happens to another person. The fervent energy behind prediction markets is fueled by its own content ecosystem, which is often backed by Polymarket and Kalshi’s marketing teams. More online engagement begets more action across platforms.

“If there is a fresh wallet, a lot of money and then the bet comes in, [like clockwork] it’s going viral,” says Dustin Gouker, a gambling and prediction market analyst and consultant. “The people behind that… It’s just engagement for them.”

Polymarket and Kalshi often try to separate themselves from traditional forms of gambling. Kalshi is regulated in the US by the Commodity Futures Trading Commission, an entity with a questionable appetite for enforcement compared to other agencies with more history (Polymarket’s core platform is not available in the US, though people can access it using VPNs). States like Washington and Arizona are not buying prediction market companies’ arguments: they have sued Kalshi, accusing it of operating illegal gambling operations running afoul of state laws. Donald Trump’s administration, meanwhile, has embraced the prediction market industry.

But criticisms that prediction markets are simply gambling ignore just how strange this extremely online iteration is. Whereas Caesars or FanDuel make money by setting the right odds on the outcome of the Super Bowl, the nature of prediction markets lets users take positions against each other rather than “the house.” Instead, Polymarket and Kalshi make their money based on trade volume, which means they’re not incentivized by outcome at all. For prediction market platforms, the reality of the world has no bearing.

Social platforms like X, along with private chatrooms on Discord and Telegram, have become a nexus for prediction market users to gather and discuss the industry. There is a lot of discourse, not all of it trustworthy.

“There are very intelligent people in the prediction market community, but I would say at least two-thirds of the content is kind of crap,” says Aaron Courtney, a Kalshi user who along with his brother also runs Kalshinomics, an analytics platform. “You have to filter through the signal and noise, and a lot of it is just hype, because that gets engagement, and to some degree, it helps the exchanges.”

Elisabeth Diana, a spokesperson for Kalshi, pushed back on the posts touting potential insider trades on Polymarket that have become common on X.

“We clearly want to clear up the confusion around us and Polymarket,” Diana says. “We ban insider trading. Polymarket does not. We want to make it clear to people that we do not promote insider trading.” Polymarket didn’t respond to The Verge’s request for comment, but in recent weeks has introduced restrictions on using information that would “violate a preexisting duty or obligation of trust.”

The exchanges themselves — and their employees — are very active on X. Polymarket’s official X account, for example, regularly shares misleading or totally inaccurate information, dressed up to resemble news media by starting posts with “JUST IN” or “BREAKING.” It’s in prediction markets’ own interest to drum up activity via panic-inducing social media posts: Polymarket and Kalshi make money by charging fees when users make trades; they benefit when there’s more trade volume, and every tweet with “breaking” news has the potential to get people to move money.

Both Polymarket and Kalshi have brought on armies of X accounts to share prediction market content through influencer programs, wherein users get a company icon badge next to their name, along with access to a paid X subscription. In the past, some of those accounts have pretended to be journalists, shared false information, and posted antisemitic content. At one point, Kalshi gave a 15-year-old an X badge; he was eventually removed from the program, with The Wall Street Journal reporting that a Kalshi employee said to him, “Yo brother, legal team confirmed that we can’t work with minors rn. Kinda sad tbh.”

The expectation is that influencers bring the exchanges into their regular content. Courtney told The Verge that he has had both a Polymarket and Kalshi badge at one point or another — and has had both badges on X pulled from his account. In one incident in early February, Courtney posted on X lightly joking about Polymarket and Kalshi both giving out free groceries in New York around the same time, with the implication that Polymarket had one-upped Kalshi. Courtney says he was told he was not “Kalshi-aligned” enough and lost his badge. At another point, Courtney received a Polymarket badge because he was building tools for the platform; he said he lost that badge after posting flattering things about Kalshi.

“My thing is I like to have unbiased takes,” Courtney says, “but I have been an affiliate of both and have been dropped by both exchanges because I did not shill them, I guess.” (Diana, the Kalshi spokesperson, said the company has “policies” around what badged accounts can post, and accounts not adhering to rules contributed to the company ending the program.)

Kalshi pulled its badges from X accounts in February, but The Verge found several big accounts that indicate they are Kalshi “partners,” meaning they are paid to post about the platform. Whale Insider, which describes itself as a “leading source for non-biased crypto, tech, finance, economic, and world news,” says it is a Kalshi partner in its bio. World of Statistics, an account with five million X followers, and “Walter Bloomberg,” a popular breaking news aggregator account, are also Kalshi partners according to their bios.

Undisclosed paid content has been an issue on X: the company’s head of product Nikita Bier has posted multiple times about it. X only added the ability for users to label posts as paid partnerships in early March. The Verge asked Diana about posts from Kalshi partner accounts shared before the new X feature was rolled out that appeared to be paid content but had no such disclosure. Diana said the company has disclosure rules for partners but that some had not abided by them in the past. The Federal Trade Commission requires that content creators disclose when content is part of a paid partnership with brands.

Polymarket and Kalshi have launched a full court press of marketing, PR, and advertising to embed themselves into the daily lives of millions of people. The Associated Press announced it would license its election data to Kalshi. Substack and Polymarket have partnered to inject prediction market data into the most popular newsletter — and writers not in the early testing group are getting separate offers from exchanges to get paid to mention and cite prediction market data.

“This first year of the explosion of prediction markets, it’s been a lot of getting people talking about you and dominating the conversation. [Kalshi and Polymarket have] succeeded in that in a lot of ways,” Gouker says. “Nobody was talking about them a year ago. Now they are in media organizations. They are cited all the time… They’re very good at getting people to talk about them.”

In late March, Polymarket announced it was launching a referral program to all users who had traded at least $10,000 on the platform, allowing them to get kickbacks when other people start on the exchange with their invite link. Most of these new bettors will likely lose money like their progenitors — the top of the pyramid needs a base beneath it. What does Polymarket care, if it gets paid either way?

Stories of potential insider trading on prediction markets have effectively become a genre of their own. The goal becomes finding those accounts before the event happens — not to stop them, but to get in on the action with them. You, too, could piggyback off of inside information.

Tools like Insider Finder and 0xinsider try to analyze individual traders and attempt to detect suspected insiders or high-performing accounts so other users can follow — and copy — their moves.

Because there is an audience for content that claims to find insiders, there is also an opportunity for deception. Rajiv Sethi, an economist at Barnard College who has studied prediction markets for years, lays out a few possibilities: There’s spoofing, where a trader who isn’t an insider bets big in a way that makes others believe they could be. If other traders copy that move and buy “Yes” shares — effectively duped — the price of the “Yes” contract goes up while the price of “No” goes down. The original trader could then create a separate account and buy even more shares of “No” for cheap.

“The original wallet that is pretending to be an insider loses money,” Sethi says. “But because Polymarket doesn’t have a know-your-customer requirement that it enforces, you make even more money on this other wallet, and nobody knows that these two wallets — or maybe 10 wallets — are owned by the same person or entity.” (Kalshi collects far more personal information from users, including social security numbers, and prohibits owning multiple accounts).

Polymarket and Kalshi allow users to sell their positions before an event actually happens, meaning traders can make (or lose) money even before an event is concluded and the platform has “resolved” its outcome. The genre of “look at this suspected insider” content is good for social media engagement, but it’s also potentially a cash cow: If you broadcast a “suspected insider” after you’ve made a move and enough people copy the activity, the value of your shares will increase, making your position worth more. You could then cash out before the event even happens.

“That way you make a profit even without taking your risk,” Sethi says.

This tactic would work well on markets where the outcome is determined by a small number of people: Maduro’s capture, for example, or the nomination of a Supreme Court justice, Sethi says. Something like “Who will win the 2028 presidential election?” is harder to game this way, given the large number of people involved in the outcome of the event.

Interest in prediction markets has come in waves, but it is bigger than ever now, Sethi says. In the early 2000s, a project by the Department of Defense called the Policy Analysis Market (PAM) envisioned a platform where experts could place bets on events in the Middle East as a way of forecasting. The program was killed in 2003 after outrage by elected officials, who said the market could facilitate terrorists betting on attacks and then carrying them out — insider trading at a fatal scale, made possible through anonymity on the platform. John Poindexter, head of the group who had developed the concept, resigned soon after.

“What we are seeing now is his vision come to life through the crypto-based Polymarket, especially,” Sethi says. “It’s basically the Wild West.”

Recall the “suspected military insider” referenced in posts on X that won $90,000 correctly betting on military events like Maduro’s kidnapping and US forces striking Iran. That account has since sold all of their positions on the US entering Iran and dumped them when the price of the position was more than what they bought them for. The Verge can’t say definitively whether the Polymarket account in question belongs to an insider, but we can say that person definitely made just shy of $10,000.

By March 31, Benjamin Netanyahu was still the prime minister of Israel. Users still had no clarity on who dududududu22 was. They could’ve been someone with deep knowledge of the Israeli government’s inner workings or a person that had been duped by an unsubstantiated internet conspiracy theory. They could have been a troll, trying to bait others into following their moves. It could have belonged to a network of accounts, all hedging their bets. The account was down more than $170,000, but no matter — more “insiders” were popping up everyday.

“FOUND THIS SUSPICIOUS WALLET DOING IT AGAIN,” a post on X proclaimed about a set of Polymarket trades on Iran and oil prices. “What does he know that we don’t?” reads a follow-up with a link to the profile. Once again, the post is marked as a paid partnership.