Shorting Kalshi and Polymarket
Trading volume is booming, but an onslaught of lawsuits and exposés—and quiet warning signs from Google—point to real trouble for prediction markets.
One of my 26 locks for 2026 was that a massive, embarrassing scandal would finally compel Congress to regulate prediction markets. As a result, companies like Google would run from their nauseating partnerships with Kalshi and Polymarket.
This was not a bold take on my part, nor has it come true, at least in the way that I foresaw. A few congressional bills that would significantly rein in the worst excesses of prediction markets—especially sports betting, which makes up a sizable percentage of “trades” on these platforms—have not gotten much traction yet. There also hasn’t been a singular, glass-breaking scandal this year. Americans are gambling quite a bit, and the World Cup was unbelievably lucrative for both Kalshi and Polymarket. A recent poll even showed that a significant portion of Gen Z is deliberately moving investment funds from stocks to sports bets. Presumably, this has been made much easier by Robinhood, which has an app that allows users to easily toggle between investing in Apple and “trading” on “sports events.”
Despite growing revenues and a lack of congressional oversight, I’d argue the last handful of months have actually been pretty terrible for the main prediction markets, which are failing to fend off a painful death by a thousand cuts. They, and their Trump allies at the Commodity Futures Trading Commission (CFTC), seem legitimately spooked by an onslaught of lawsuits. Google seems to be spooked too, though its quiet retreat hasn’t gotten nearly as much attention as its announcement last year that it was integrating prediction market data into its AI-enhanced “Finance” tab.
Cities and states are suing prediction markets, mostly because of the sports bets that Polymarket and Kalshi actively advertise. As I’m writing this, news just broke that the city of Baltimore is suing both Kalshi and Polymarket, alleging that they’re operating as unlicensed gambling companies. (This is on top of a separate lawsuit filed by the state of Maryland, which is ongoing.) Minnesota came dangerously close to becoming the first state to ban prediction markets outright, until a federal judge issued a preliminary injunction a few weeks ago. On July 31, the state of New York sued Kalshi; State Attorney General Letitia James said in a statement, “No matter what they call themselves, prediction markets like Kalshi are gambling platforms, plain and simple.” New York City Council has since announced a separate investigation into “deceptive and predatory marketing practices by prediction market platforms,” according to a press release.
The cherry on top: 44 state attorneys general—forty-four!—complained to the CFTC last month that it is exceeding its mandate by attempting to regulate sports betting on prediction markets. Above all else—the addictive nature of gambling, safeguards for minors, the societal rot of being able to wager on basically everything—blue and red states alike are displeased that they aren’t collecting taxes on prediction market bets placed within their boundaries. They are right to be displeased! For all of the sportsbooks’ many faults, they are, at least, taxed on their revenues by states that have chosen to permit DraftKings, FanDuel, etc.
Kalshi and Polymarket have mostly hidden behind the CFTC and a defense that they are federally licensed exchanges not beholden to individual states. Like most Trump-friendly entities, the CFTC isn’t very convincing. On Aug. 11, the agency exercised its “emergency authority” to keep Kalshi operating in New York. This was done in the spirit of “market stability,” as if the entire Kalshi ecosystem would collapse because, during the dog days of summer, guys in Queens could no longer place “event contracts” on Da Freakin New York Mets. I am only barely kidding: in a statement, CFTC Chairman Michael Selig referred to Kalshi as one of multiple “interstate financial markets,” dramatically adding, “New York intends to make event contract derivatives waste away under its iron curtain of state gaming laws before the courts get the chance to issue final rulings.”
With the CFTC’s protection, prediction markets are going to squeeze as much juice as they can out of sports bets, as well as novel forms of degeneracy like gambling on flight delays. (FlightAware actually sued Kalshi this week for using its data and trademarks, then withdrew the suit, perhaps because Kalshi removed references to FlightAware and added disclaimers about its data.) I’m sure there will be excitable headlines this fall about all the money that prediction markets are racking in from NFL games and midterm elections.
But look beyond the betting volume. The lawsuits speak for themselves; they’re existential. The CFTC, which is a lackey for the prediction markets so long as Trump is in office, is interspersing dramatic statements about a gambling “iron curtain” with (mealy-mouthed, but still!) concessions meant to eliminate some of the stupidest “event contracts.” A Wall Street Journal exposé about false, deceptive advertising from Polymarket was damaging enough to compel a supposed company audit, plus an anonymous admission from the CFTC that it’s conducting an investigation. In response to New York City Council’s probe into the marketing practices of prediction markets, Polymarket issued a surprisingly conciliatory statement: “We look forward to engaging with The New York City Council on this matter,” a spokesperson told CNBC.
You can also intuit Kalshi and Polymarket’s precarious position through Google’s lawyer-speak and the fact that its prediction market integration has been a dud. We’ve come a long way since November 2025, when the illegal search monopoly slash journalism-killer touted a new feature where users could open up the Google Finance tab, “ask questions about future market events” (referring to Kalshi and Polymarket), and “harness the wisdom of the crowds.”
In April, Futurism reported on how Polymarket bets were showing up in the Google News tab next to actual news stories. Google issued a statement chalking up the Polymarket promo as an error. Then, sometime between June and July, Google added a skittish disclaimer about its use of prediction markets data. “This data is sourced directly from third-party market providers and is not generated or calculated by Google,” the disclaimer reads. It also includes a section on “regulatory and financial risks,” including that “the regulatory framework surrounding prediction markets is complex and continues to evolve,” so “you are responsible for understanding and adhering to the regulations applicable in your specific location.”
That’s not all! As of August 1, the Google Chrome Web Store outright banned extensions that “facilitate or enable real-money transactions on predictive outcomes.” The Chrome Web Store didn’t differentiate between sportsbooks and prediction markets: “We don’t allow content or services that facilitate or promote real money gambling or prediction markets, including but not limited to online casinos, sports betting, lotteries, or games of skill that offer prizes of cash or other value.”
Last night, I messed around with Google Finance’s main search tab, the real hub for this prediction market partnership. I was hoping to pull up some sports and midterm election “event contracts.” Wouldn’t you know it? No results. I was only able to surface forecast modules for “event contracts” outside of sports, politics, and culture. I tried to chat with Gemini about this, which repeatedly returned error messages and gave conflicting information about why the Finance tab wasn’t prominently displaying sports or election events via Kalshi and Polymarket.
Best as I can tell, Google has been filtering out controversial sports and election markets from its Finance tab. It probably has been doing so for a while, if not the entirety of their partnership. On one hand: good! Makes sense! On the other: not a ringing endorsement of the primary use case for Polymarket and Kalshi.
As cities, states, and media outlets continue to sue and scrutinize prediction markets, I suspect Google and other initially proud allies will further distance themselves, so as to avoid additional reputational harm. Including, one can only hope, Kalshi’s top talent: Timothée Chalamet. What the hell, man?
Here’s what else we’re reading this week:
Wired and Mission Local reported on two different AI-run websites that purport to be journalistic ventures.
One of the outlets, RuntimeWire, has actually scored a few scoops by ingesting transcripts and court documents, then publishing news stories, faster than humans can. Most of the articles I scanned on RuntimeWire are bland and poorly structured, and the site’s creator doesn’t exactly have a foolproof plan for editing, fact-checking, and legal reads. (In short: he uses AI agents and his own judgment.) But I can at least see the usefulness of having AI tools constantly scanning for newsworthy transcripts and filings.
The other AI-run site highlighted by Wired and Mission Local is The Dissent. I do not like it. Every story has a fake byline, so as to assign an anthropomorphic quality to the AI agents who are (badly) aggregating actual journalism. The creator of The Dissent did not consult with any journalists about this pet project, according to Mission Local.
Xavier Becerra has been quiet since clinching the Democratic Party nomination for California governor. He made a rare public appearance at a Politico-sponsored event earlier this week, and demonstrated why others in his party do not speak very highly of his message discipline and political vision.
Becerra floated something or other about offering shares in AI companies to Californians, then walked it back, while adding that such an idea is, ultimately, not “totally far fetched.” So which is it? I suppose we’ll find out eventually.
Becerra also lamented how “we hardly have any” AI regulations in California, which is true, though California still far outpaces other states.
And Becerra stepped on the toes of Rob Bonta, who succeeded him as state attorney general, by suggesting California’s lawsuit to block the merger between Paramount Skydance and Warner Bros. should end with a settlement.
Rep. Ro Khanna, whose district encompasses Silicon Valley, seems to be inching closer and closer to an official presidential run. Khanna recently swatted down a Big Tech-backed challenger in his primary, and told NPR that achieving “America’s next New Deal” will require standing up to tech oligarchs.
“These folks would have been the conquerors in a different era,” he said. “They kind of think of themselves as above you and I. And when I’m actually most inclined to run, I think, ‘I know these people.’ I understand the economic future, and I will have the guts to stand up to them.”
Sam Altman told a group of tech interns the following: “I think we are close to a world where you can have, like, a descendant of ChatGPT watch your computer screen all of the time, watch every meeting you’re in, record every call, everything like that. Have perfect context of your whole life, everything you see. You choose what information you want it to have, but it can like—you can connect it to your texts or emails or Docs or Slack or whatever. And then you kind of have this thing that’s not making decisions for you, but if you’re the CEO of a startup, there’s always more stuff for you to do and context you can’t all keep track of. You can’t read every piece of customer feedback every day. You can just have this thing working alongside you.”