An ultimatum, delivered from a fintech stage
A conference panel on financial technology is an unusual venue for a gambling-policy ultimatum, which is roughly what Kalshi’s chief executive delivered in Washington. Speaking at DC Fintech Week, Tarek Mansour warned that prediction market operators could move their businesses outside the United States if courts uphold states’ authority to regulate event contracts as gambling, as reported by Focus Gaming News on 15 October 2026. That single remark is the clearest public statement yet of what is genuinely at stake in the fight over prediction markets regulation: not which agency writes the rulebook, but whether the product stays within reach of any American regulator at all.
A prediction market is an exchange where people buy and sell contracts that pay out based on whether a real-world event happens. The question underneath Mansour’s warning is whether that sentence describes a derivative or a bet. American law currently gives two different answers, depending on which courthouse you walk into.
What is a prediction market, exactly?
A prediction market is a venue for trading event contracts: standardised, binary instruments that settle at a fixed value if a stated outcome occurs and at zero if it does not. Prices move between those two poles and function as implied probabilities. Participants trade with each other through an order book, not against the house.
The mechanics are worth spelling out, because they are where the regulatory argument actually lives. A contract on “Team A wins” priced at 60 cents settles at one dollar if Team A wins. Buy it and you have paid 60 cents for a dollar of contingent value, which prices the outcome at roughly 60%. Someone else takes the other side at 40 cents. The exchange matches you and takes a trading fee. It has no position and no interest in which side is right.
Prediction market mechanics versus a sportsbook bet
A sportsbook does have a position, and prices accordingly. Two sides of a market at -110 each imply 52.4% probability apiece, which sums to about 105%. That extra five points is the overround, the margin baked into the price. You cannot trade out of it, you cannot sell your ticket to another customer at a better number, and the counterparty to your bet is the operator itself.
The practical differences follow from that. On an exchange you can exit a position before settlement at whatever the market will pay, which means a contract bought at 60 cents and sold at 80 cents realises a gain without the event ever resolving. With a fixed-odds bet, the slip is settled by the result, full stop. One is a tradable instrument with a secondary market; the other is a wager with a fixed payout. Whether that distinction is substantive or cosmetic is, more or less, the entire lawsuit.
What provoked the warning
Mansour was defending Kalshi’s position in a running legal conflict over whether event contracts fall under federal commodities law or state gambling rules. State gambling regulators have asserted jurisdiction over sports-related event contracts; Kalshi, as a federally regulated exchange, argues the Commodity Exchange Act and the CFTC pre-empt them. Courts have not landed uniformly, and the Focus Gaming News report notes Kalshi has taken setbacks in some of them.
His framing was deliberate. “It’s not a choice of state versus federal market. It’s a choice of whether prediction markets exist in the US or whether they go offshore,” he said, while adding that he felt “pretty good” about the company’s legal position. Read it as a forecast and it is speculative. Read it as a description of operator incentives under fifty separate licensing regimes and it is closer to arithmetic.
One caveat on tone: this is an executive making a case in public, not a neutral assessment of where the law is heading. Nothing he said is a commitment to leave, and nothing here should be read as a prediction about what Kalshi will actually do.
Two products, two rulebooks
The structural gap between a CFTC-regulated exchange and a state-licensed sportsbook is wider than the customer experience suggests. Both can be reached from a phone in under a minute. What sits behind the app differs substantially.
| Feature | Prediction market (event contracts) | Licensed sportsbook (fixed odds) |
|---|---|---|
| Primary regulator | CFTC, as a federally designated exchange | State gaming commission or control board, per state |
| Product structure | Binary contract traded between participants | Wager against the operator at posted odds |
| Counterparty | Other traders, cleared through the exchange | The operator itself |
| Revenue model | Trading and settlement fees on volume | Margin built into the odds (the vig) |
| Exit before resolution | Sell the position at market price | Cash-out only where the operator offers it |
| Dispute path | Exchange procedures, then federal complaint channels | Operator complaints, then the state regulator |
| Geographic scope | Nationwide under federal registration | Only states where the operator is licensed |
Fee schedules and fund-handling requirements differ by venue and change over time; check the exchange’s published fee page and the CFTC for current federal requirements rather than relying on a summary table.
The column that matters most to a consumer is the last two. A state gaming board is a body with a phone number, a complaints process and licensing leverage over the operator it regulates. Federal commodities oversight exists and has teeth, but it was not built around retail customers disputing the settlement of a football market. Neither framework was designed for this product, which is precisely why the jurisdictional fight has lasted as long as it has.
Why regulatory pressure pushes platforms offshore
Gambling-adjacent platforms relocate offshore when the cost of compliance exceeds the cost of losing access to a market, and that calculation tips earlier than most people assume. It is not usually one dramatic decision. It is a sequence of smaller ones.
- Licensing multiplies. A single federal registration becomes separate applications, bonds, fees and technical standards in every state that claims authority.
- Legal ambiguity raises the discount rate on everything. A platform that cannot say with confidence whether it is lawful in a given state struggles to raise capital, sign banking partners or plan a product roadmap.
- Enforcement risk is asymmetric. Being wrong about jurisdiction can mean cease-and-desist orders, frozen accounts or criminal exposure for individuals, while being right earns nothing extra.
- Lighter-touch jurisdictions compete for the business. Offshore licensing regimes market themselves on speed, cost and permissiveness, and they are reachable by any customer with an internet connection.
Online gambling has already run this pattern twice. Offshore sportsbooks served American customers for decades from jurisdictions with minimal oversight, and more recently crypto-denominated casinos have clustered in light-touch licensing hubs while remaining accessible almost everywhere. Specific operators, dates and licensing jurisdictions in that history should be sourced to primary regulatory records or reputable trade reporting rather than taken on assertion, including anything that looks like a tidy statistic about offshore market share.
The uncomfortable part is that regulatory arbitrage is a rational response to a genuine problem. Fragmented, contradictory rules impose real costs on operators trying to comply. Pointing that out is not an endorsement of leaving, and an operator’s incentive to go is not the same as a consumer’s interest in it going.
What offshore has meant for players before
When a platform moves beyond domestic regulation, the protections that disappear are mostly invisible until something goes wrong. The product on screen looks identical. The recourse behind it does not.
Four things typically weaken at once. Licensing oversight thins, because a regulator with no local enforcement power cannot compel much. Dispute resolution loses its backstop, so the only appeal from an operator’s decision is the operator. Fund segregation and solvency arrangements become opaque, meaning customer balances may sit on the company’s own books rather than in protected accounts. And KYC and AML obligations are enforced to whatever standard the host jurisdiction chooses, which is sometimes rigorous and sometimes nominal.
Against that, the honest counterweight: plenty of offshore-licensed platforms have operated for years without stiffing customers, and domestic licensing has never been a guarantee of good behaviour either. The difference is not that offshore operators are necessarily worse. It is that when they are worse, there is no one to call. Any specific historical example of frozen balances or an offshore collapse should carry a citation to court filings, regulatory notices or established reporting before it appears in print.
This applies to prediction markets with one additional wrinkle. Because event contracts are framed as trading rather than betting, the consumer-protection tooling that gambling regulators have built up over a decade, including deposit limits, self-exclusion and reality checks, is not uniformly embedded in exchange rulebooks. Trading an outcome you cannot influence carries the same financial risk as betting on it, whatever the product is called, and the same discipline about stake sizing and limits applies.
The question nobody has answered yet
The real fork is whether American lawmakers build a bespoke framework for event contracts or keep litigating the question until the answer is decided somewhere else. A purpose-built regime could borrow the market structure of an exchange and the player-protection furniture of gambling regulation, which is a sensible combination that nobody currently has a mandate to write. The alternative is not a ban. It is the outcome Mansour described: the product continues, the customers continue, and the oversight is whatever a distant licensing authority decides it should be.
Which is why the detail to watch is not the next court ruling in isolation but whether any legislative vehicle emerges alongside it. Litigation can settle who has jurisdiction. It cannot write the rulebook that would make the answer worth having.
