On 9 October, Kalshi opened an investigation into its own customers. Someone, or several someones, had placed trades correctly predicting that Katie Zacharia, a Department of Homeland Security spokesperson with no public buzz behind her, would become the next White House press secretary. The contract moved hours before Axios broke the appointment. Nobody inside Kalshi thinks this was a lucky guess. The company that built a regulated betting exchange for things like interest rate decisions and government shutdowns now finds itself doing something it never advertised: reading its own order book for evidence of who had early access to a White House personnel decision and traded on it. That is not what prediction markets were pitched as. They were sold as forecasting tools, a way to crowdsource the future by letting people put money where their mouth is. What they are turning into, almost by accident, is a tripwire for leaks. Institutions that have spent decades assuming their internal decisions were safe from scrutiny are discovering that a few thousand dollars on a betting app can expose them faster than a subpoena.
Start with the mechanism, because it explains everything that follows. A prediction market works by converting a question into a tradeable price. "Will Katie Zacharia be named press secretary?" becomes a contract that moves between zero and a dollar depending on how traders bet. If the price sits at four cents for weeks and then jumps to eighty cents in the hours before an announcement, that jump is not noise.
It is a signal that somebody with better information than the market has decided to act on it. Equity markets have policed this exact pattern for a century. Unusual volume before a merger announcement, options activity that spikes the day before earnings, these are the tells that send compliance officers digging through trade logs. Kalshi's press secretary contract did the same thing, except the leak wasn't coming out of a boardroom on Wall Street.
It was coming out of the West Wing. That distinction matters because nobody built Kalshi, or Polymarket, or any of the dozen platforms now trading on political appointments, legal verdicts and corporate outcomes, to catch leakers. They were built to let people bet on the news. The Commodity Futures Trading Commission regulates Kalshi as a designated contract market, the same category as agricultural futures exchanges, which means it has real obligations around market integrity but none of the decades of insider-trading jurisprudence that governs stocks.
Nobody at the CFTC wrote a rule anticipating that a prediction market would become the fastest public instrument for detecting leaks inside a presidential administration. The rule didn't need to exist, because nobody expected the markets to get big enough, or liquid enough, for the signal to show up this cleanly. It is showing up now because the money got serious.
Bloomberg's analysis of Kalshi and Polymarket trading, published 7 October, found that longshot bets losing 98% of the time still dominate volume on both platforms, which tells you plenty of the activity is still closer to a casino than a research tool. But underneath that noise sits a growing pool of traders who are not gambling, they are informed, and their trades are large enough and timed tightly enough to be legible.
The Financial Times made the broader case the same week: trades on these platforms increasingly contain genuine signal for investors and executives willing to read them, not because the crowd is wise in some mystical sense, but because specific individuals inside specific rooms know things before the public does, and now have a frictionless way to monetise that knowledge in real time. , - The press secretary episode is small beer next to what this implies for corporate and financial information.
Polymarket has been taking bets on whether HSBC and Lloyds will fail, a market that prompted UK politicians to demand regulatory intervention this month, according to the Guardian's 3 October report. Think about what that contract actually is. It is a public, dollar-denominated price on the solvency of two systemically important banks, trading on a platform outside the Bank of England's supervision and outside the Financial Conduct Authority's remit, because Polymarket operates offshore and markets itself as event contracts rather than financial instruments.
If someone with privileged knowledge of either bank's liquidity position decided to take a position ahead of bad news, the trade would be visible to anyone watching the order book, and invisible to the one regulator with the authority to ask who placed it. That is the structural problem nobody has solved yet. A hedge fund analyst with material non-public information about a UK bank cannot legally trade that bank's shares on the London Stock Exchange without triggering an insider-dealing investigation under the Financial Services and Markets Act.
The same analyst can, as of today, buy a Polymarket contract on whether that bank fails, in a market with none of FSMA's reporting requirements, none of the LSE's surveillance infrastructure, and settlement in crypto that makes identity harder to trace than a brokerage account. The UK has no mechanism to compel Polymarket to hand over trading records the way the FCA can compel a London broker.
Robinhood and Kalshi at least sit inside US jurisdiction and have been cooperating, however reluctantly, with CFTC oversight. Offshore platforms do not have to. "Trades can contain valuable cues for businesspeople and investors," the FT noted, which is a polite way of saying the same information asymmetry that insider-trading law exists to punish on regulated exchanges is now available, legally, to anyone clever enough to trade it on an unregulated one.
Blanche Lincoln, the former US senator who once wrote the legislation banning sports-event contracts, is now a paid lobbyist for Kalshi, a detail that tells you how completely the political consensus on these markets has inverted in under a decade. The industry that regulators tried to ban as gambling is now being courted as a forecasting tool too useful to shut down, even as it quietly accumulates the exact surveillance liability that equity markets spent a century building rules around. , - The counterargument deserves a fair hearing, because it is not a weak one.
Prediction markets are thin. Kalshi's press secretary contract probably had a few hundred thousand dollars in volume, nothing like the billions that trade through options markets ahead of an earnings call. A handful of informed trades on a shallow market can move the price dramatically without requiring any actual leak at all, just a well-connected guesser with conviction and a few thousand dollars of risk appetite.
Correlation between a price jump and a later-confirmed outcome is not proof of inside information, it might just be proof that some trader reads Politico more carefully than everyone else. Kalshi's own investigation into the Zacharia trades had not produced a public conclusion as of this month, and it may well land on "educated speculation" rather than "leak." But the direction of travel is not in question even if individual cases stay ambiguous.
Every month these markets grow, the sample size of suspicious trades grows with them, and the pattern-matching gets easier for anyone watching, whether that is journalists, short sellers, or the compliance teams inside the institutions being bet on. The Segantii Capital Management trial that wrapped up in Hong Kong on 9 October, with a verdict due in February, is a reminder of how long and expensive traditional insider-trading enforcement actually is: years of investigation, a high-profile criminal trial, uncertain outcome even with full subpoena power and a dedicated financial crimes unit.
A prediction market throws up the same signal in hours, for free, visible to anyone with a browser. That is a startling asymmetry between the cost of detection and the cost of prosecution, and it is going to shape who gets caught first. , - The practical shift for anyone running a company, a campaign, or a government press office is that the assumption of a sealed decision no longer holds once a liquid market exists on its outcome.
If there is a Kalshi or Polymarket contract on your next CEO announcement, your merger approval, your regulatory ruling, treat the order book the way a listed company treats its own share price before an acquisition: as a leading indicator that someone, somewhere, already knows more than they should. Boards making sensitive appointments should assume the betting markets are watching before the press is, because increasingly they are.
Investors reading these markets for signal should separate genuine information from thin-volume noise, since Bloomberg's data shows the bulk of activity is still retail punters chasing longshots rather than insiders trading on certainty. Regulators face the harder question. The NFL's request to the Supreme Court this month, asking the justices to let states regulate prediction markets despite the Trump administration's preference for federal CFTC oversight, is nominally about sports betting jurisdiction.
It is really about who gets to decide whether these platforms need the surveillance infrastructure of a real exchange. Right now nobody has built that infrastructure, which means the next leak these markets expose, inside a bank, a pharma trial, a court verdict, will be caught by accident, by a journalist or a short seller noticing a weird price move, rather than by design.
That is a strange way to run financial surveillance, but it is the one currently in place, and it is catching more than anyone intended.
Briefed+
Sunday is the paid one.
Around 2,000 words on the one thing that actually moved this week, written for people who have to make a decision about it. Included in Briefed+, or earned by sending three people to the free Daily.