On October 14, five days after the Commodity Futures Trading Commission filed a proposed rule that would formally categorize event contracts as "swaps" while explicitly carving sports and casino wagering out of federal jurisdiction, Kalshi CEO Tarek Mansour walked onto the DC Fintech Week stage and did something subtler than a victory lap. He changed the question. He did not argue about whether prediction markets are legal. He argued about whether they will exist onshore or offshore — a binary he repeated, deliberately, like a mantra. If American platforms cannot serve American users, he implied, those users do not stop betting. They just route their dollars to a server in Singapore, or Zug, or wherever the law has not yet decided to care.
That is the tell. When a founder reframes a constitutional dispute about state gambling law into a simple matter of national competitiveness, he is not explaining a market. He is manufacturing one. And I have watched this exact choreography before — the moment a regulated actor stops defending its product and starts auctioning the definition of the product to the highest jurisdictional bidder.
The Question Nobody Asked in 2021
To understand why a single CEO's framing matters, you have to understand what a prediction market actually is — and why regulators have spent a decade refusing to name it.
A prediction market is a venue where participants buy and sell contracts whose payoff depends on the outcome of a future event: an election, a Fed rate decision, a hurricane's landfall, the price of Bitcoin at year-end. The contract price, in theory, functions as a probability. If a "Yes" contract on some event trades at 63 cents, the market is collectively whispering that the event has a 63% chance of happening. This is not a gambling product dressed in a suit. It is an information aggregation mechanism — a machine for converting dispersed, private belief into a public, tradeable number. The academic literature on this is older than most of the tokens I have audited; prediction markets have been studied as forecasting tools since the late 1980s, and their track record on elections has, at times, outperformed polling averages.
But here is the cultural friction that no whitepaper resolves: to the average state regulator, a contract that pays out when a football team wins is not an information instrument. It is a sportsbook. And a sportsbook, in the American legal imagination, is a vice — something governed by state gambling statutes written long before anyone conceived of a decentralized order book. The entire fight, stripped of its technical varnish, is a semantic war over which vocabulary wins: "derivative" or "wager."
Kalshi is a designated contract market, or DCM — a license granted by the CFTC that makes it a federally regulated derivatives exchange, the same category as the venues where crude oil futures and interest-rate swaps trade. Polymarket, its loudest competitor, is a crypto-native platform that settled with the CFTC in 2022 and has since leaned into on-chain transparency, a hybrid of a central limit order book and automated market making, with global reach and a US footprint that has historically been constrained. Two platforms. Two philosophies. Two entirely different answers to the question of who gets to say what a contract is.
On October 9, the CFTC answered. It proposed that event contracts be classified as swaps under the Commodity Exchange Act — placing them squarely inside its own jurisdiction — while excluding contracts tied to sports and casino gaming. Kalshi's response came five days later, timed with the precision of someone who understands that a 30-day public comment window is not a formality but a battlefield.
Reading the Code That Isn't There
Here is where my training gets impatient. I spent three months in 2017 reverse-engineering the Zeppelin security library, submitting patches, learning that the truth of a system lives in its smallest mechanical choices rather than its marketing. So when I read a regulatory transcript, I read it the way I read a smart contract: I look for what the architecture admits but the narrative hides.
Kalshi's article contains almost no technical disclosure. No TPS, no latency, no settlement mechanism, no oracle design. But it contains one forensic fingerprint that a careless reader would skim past. When Mansour was pressed on accusations that Kalshi's trading volumes were inflated, he did not deny the numbers outright. He attributed them to something specific: market makers placing fixed-size orders. Read that again, because it is a confession disguised as a rebuttal. Fixed-size market-maker orders are a hallmark of an order-book venue. They are not how an automated market maker behaves — an AMM does not "place orders," it quotes against a bonding curve. So Kalshi is running a centralized order book, matched by an engine it controls, seeded by professional liquidity providers. That is the architecture of a traditional exchange, not a DeFi protocol. And that architecture has a well-documented pathology: when the same market-maker quotes both sides at fixed size and those quotes cross and re-cross, the tape can double-count economic activity that never carried independent risk.
The volume-inflation allegation is not primarily a story about fraud. It is a story about a measurement methodology — and the fact that Kalshi has no on-chain ledger to independently falsify it is the real vulnerability. Polymarket, for all its flaws, has a public chain. Every trade is a verifiable fact. Kalshi, by virtue of its compliance-first posture, has surrendered exactly the transparency that would settle the dispute. This is the structural irony of regulated crypto: the license that grants legitimacy also removes the auditability that grants trust.
I have seen this trade before, in a different costume. In 2020, I spent weeks hunched over fifty protocol dashboards during DeFi Summer, mapping how yield farms reported TVL that was partly circular — liquidity that existed only because it was simultaneously borrowed and deposited. The number was not a lie in the accounting sense. It was a lie in the epistemic sense: it measured a quantity that no independent observer could reconstruct. Kalshi's volume dispute is the same species of ambiguity, imported into traditional finance.
The Swap That Launders a Bet
The CFTC's October rule is the hinge on which this entire story turns, and its logic deserves the same forensic scrutiny. By defining event contracts as swaps, the agency routes them through the Commodity Exchange Act rather than the securities framework. This matters enormously. A swap is a derivative in which two parties exchange cash flows contingent on some underlying reference. It is not a security — which means it does not need to satisfy the Howey test, the four-pronged standard (investment of money, common enterprise, expectation of profit, reliance on others' efforts) that determines whether an instrument falls under SEC purview.
Run the Howey test against an event contract and the seams show. Money is invested — yes, plainly. Is there a common enterprise? Arguably not; the payoff depends on an external event, not on a promoter's managerial effort. Is there an expectation of profit? For a speculator, yes. But is that profit derived from the efforts of others? Almost never — it is derived from whether the world turns out a certain way. The event contract fails Howey on the prongs that matter, which is precisely why the CFTC's "swap" designation is a jurisdictional masterstroke: it lets the federal government claim the product without ever inviting the SEC to the table.
But notice the exemption buried inside the rule. Sports and casino contracts are excluded. This is not a technical carve-out; it is a political concession, and its fingerprints are all over the lobbying record. The major sports leagues — the NFL most visibly — have fought prediction markets on sports outcomes with the ferocity of incumbents protecting a franchise. The exclusion is what allowed the rule to survive internal politics, and it simultaneously caps the addressable market of every onshore prediction platform. Kalshi gets federal legitimacy, but it does not get the one category of event that generates the most liquidity: the game on Sunday.
That is a strange bargain. It is also a revealing one. It tells you that the CFTC is not trying to legalize gambling. It is trying to legalize information markets while amputating the limb that looks most like a casino.
The Jurisdiction Map Nobody Can Draw
The constitutional picture is messier than any headline suggests. The federal appeals courts have split: at least one ruling has favored Kalshi's federal preemption argument, while two have sided with the states. New Jersey has petitioned the Supreme Court to hear its case, and the leagues have signaled support for the state position. What this produces is not a settled regime but a legal uncertainty premium — a tax on every participant in the sector, priced in the currency of litigation risk rather than basis points.
Mansour's rhetorical move — onshore versus offshore — is a strategy to collapse this complexity into a single, emotionally legible choice. And it works, because it exploits a real asymmetry. If the states win, the geography of the product does not disappear; it migrates. A user in Ohio who cannot legally trade a Kalshi contract will not stop wanting the exposure. They will open a VPN and a Polymarket account, and the tax revenue, the regulatory oversight, and the consumer-protection apparatus all evaporate at once. This is the strongest argument in Kalshi's arsenal, and it is also the one that conveniently ignores the counterfactual: offshore markets have their own costs — opacity, manipulation, no recourse.
The honest framing is not "onshore versus offshore." It is "which set of institutions absorbs the risk." A federal DCM absorbs it with surveillance and capital rules. A state gaming regulator absorbs it with licensing and, historically, with a deep suspicion of anything that smells like a derivative. An offshore protocol absorbs it by pretending it does not exist. Three absorbers, three failure modes. The CEO's binary erases the middle.
The Oracle Is a Person
Now the part of the story that the compliance narrative most wants you to skip: the insider trading cases. The article documents enforcement against two individuals — Santos and Perez — for trading on non-public information about event outcomes, resulting in fines, disgorgement, and multi-year bans. On its surface, this is a success story: the platform detected misconduct, the regulator prosecuted, the framework worked.
Look closer, and it becomes a confession about where the truth of a prediction market actually lives. An event contract has to resolve. Somebody — or something — must decide whether the event occurred. In a decentralized design, that is the oracle problem, the hardest unsolved problem in the entire industry: how do you feed external reality into a deterministic machine without introducing a trusted party who can lie? Kalshi does not solve this. It centralizes it. The platform adjudicates outcomes, and the enforcement cases prove that the platform also surveils its own users to catch those who trade on foreknowledge of the adjudication.
This is not a criticism of Kalshi. It is a description of what a regulated prediction market necessarily is. A prediction market with a centralized resolver is not a market at all in the pure sense — it is a market plus a referee, and the referee's integrity is the product's real underlying asset. When Santos traded on inside information, he was not exploiting a bug in a smart contract. He was exploiting the gap between what he knew and what the referee would eventually announce. That gap exists in every design where a human decides the answer.
I learned to look for this in 2021, when I interviewed twenty-two community leaders for a newsletter about the cultural semiotics of digital collectibles. The floor price of an NFT was never just supply and demand. It was a social fact — a consensus maintained by a community's shared belief about who was in and who was out. Code speaks, but culture listens. The same is true here. The price of a Kalshi contract is not a pure probability. It is a probability filtered through trust in the resolver, and trust in the resolver is a cultural variable, not a technical one.
The insider trading cases, then, are not an aberration. They are the natural expression of a system where the settlement source is a gatekeeper. And here is the uncomfortable implication: as prediction markets scale, the value of insider information scales with them. The more events a platform lists, the more people exist who know the answer before the market does. Surveillance becomes not a feature but a permanent cost of doing business. The referee can never stop watching the players.
The Cold-Start Problem Nobody Prices In
There is a deeper structural flaw that neither the CFTC rule nor the CEO's framing addresses, and it is the one I would flag in any audit of the sector: liquidity fragmentation. Every event is its own market. An election contract and a hurricane contract share a platform but not a book. There is no shared pool from which liquidity can be drawn. This means that a prediction market's total value is not the sum of its volume — it is the sum of hundreds of thin, independently bootstrapped order books, each of which must find its own market makers, each of which can collapse when a single large trader exits.
Compare this to a perpetual futures market, where Kalshi also operates, and the contrast is stark. A BTC perpetual has one book, enormous depth, and continuous two-sided flow. An event contract has a book that lives for weeks and dies on settlement. The economics are almost inverted. Prediction markets are structurally hostile to liquidity, and the only remedy — market makers willing to quote thin, episodic markets — is exactly the mechanism that produced the volume-counting controversy in the first place. The complaint and the solution are the same object.
This is why I suspect Kalshi's crypto perpetual business is not a side project but a hedge. The perpetual gives the platform a high-frequency, deep-liquidity product to subsidize the episodic, thin-liquidity event business. It is a cross-subsidy disguised as diversification, and it drags Kalshi directly into the most competitive red ocean in the entire asset class — the perpetual futures market, where it will face venues with years of order-flow advantage.
The Blind Spot in the Onshore-Offshore Story
Here is the contrarian reading, and it is the one I keep returning to as I map this system's risk surface.
Everyone — Kalshi, the CFTC, the journalists, the crypto commentariat — is treating this as a contest between a compliant onshore platform and a decentralized offshore one. That framing is comfortable because it flatters both sides: Kalshi gets to be the responsible adult, Polymarket gets to be the freedom fighter. But the real competitive threat to prediction markets is not each other. It is DraftKings. It is FanDuel. It is the regulated sportsbook industry, which already owns the customer relationships, the payment rails, and the state-level lobbying infrastructure that prediction markets are only now trying to build. If the sports-betting exclusion holds, the prediction market's most lucrative category is permanently ceded to incumbents who never had to argue about whether a wager is a swap.
And the deeper blind spot is this: the entire onshore-versus-offshore frame presupposes that the product is worth fighting over. But a prediction market's social value — its claim to being an information instrument rather than a vice — depends on the market being deep enough to produce a meaningful price. Thin, fragmented, insider-contaminated books do not produce good information. They produce noisy numbers that occasionally look prescient. The Cassandra complex is real: I have spent years warning that the products crypto most wants to legitimize are frequently the ones least ready for the scrutiny that legitimacy brings. Prediction markets may be the purest example yet. The moment they get the federal blessing they have fought for, they will be judged not on their ideology but on their liquidity — and on that metric, the offshore protocol and the onshore DCM may both be found wanting.
What to Watch When the Comment Window Closes
The 30-day comment period is not the end of this story; it is the opening of a longer, slower one. The signals that matter are not the speeches. They are the filings. Watch whether the CFTC's rule survives the inevitable legal challenge — if it is struck down, Kalshi's compliance moat dissolves overnight and its whole strategic posture becomes a liability. Watch whether the Supreme Court accepts New Jersey's petition, because a ruling there will set the ceiling for every onshore prediction market in the country. Watch for the first independent audit of Kalshi's perpetual volumes, because a single credible third party with access to the order flow could either vindicate the platform or end its market-position narrative.
And watch, most of all, whether the industry ever names its real adversary. As long as prediction markets define their enemy as the offshore protocol, they will keep winning sympathetic headlines. The day they admit that their true competitor is the sportsbook down the street, the story changes entirely — from a fight over freedom to a fight over market share. That is the story the next regulatory cycle will tell, whether anyone on that DC stage wants to hear it or not.