Mansi Birla starts with what the CFTC oversees. The joint SEC and CFTC interpretation from March named Bitcoin, Ether, and several other assets as digital commodities. The CFTC regulates derivatives on these assets, perps included, plus certain leveraged retail transactions. It can also act against fraud and manipulation in spot markets. Without new legislation, it cannot make every spot exchange register with it. Mansi, who helped a US-listed crypto exchange obtain its DCM license, stresses that a CFTC license is not one license. The exchange, the clearing house, and the intermediaries that carry customer accounts are separate regulated functions. Each business decides which layers to own and which to outsource.
With the Clarity Act stalled in the Senate, the CFTC is moving through rulemaking. Regulation CTX would clarify which retail crypto transactions fall under its existing authority over leveraged, margined, or financed trades. Regulation CAM would set a tailored framework for the exchanges that offer them. Mansi Birla notes that both are at the advance-notice stage, with no proposed rule text. The preliminary design pairs an exchange with clearing arrangements and a futures commission merchant that carries customers. Mansi says institutions weigh durability as heavily as permission: a license is worth less when its operating conditions can change without warning.
The CFTC gave existing DCMs no-action relief to convert dated futures into perpetual futures, Mansi Birla explains. Only existing players qualify. She names pricing as the main obstacle to a full perps framework. Whether perps are swaps or futures is still before the courts. Perps work well in crypto because the spot market trades around the clock and funding payments keep the two prices aligned. On traditional assets, the risk changes when a thin market triggers automated margin calls and liquidations. Mansi expects deeper liquidity, and the move by Nasdaq and the NYSE toward 24/7 trading, to reduce that risk.
Hyperliquid isn't a DCM, an intermediary, or a clearing house. Asked where it fits, Mansi Birla points to Payward, Kraken's parent, which has announced plans to bring Hyperliquid onshore through its Bitnomial and NinjaTrader infrastructure. A regulated US venue has several routes to Hyperliquid's liquidity. It can use Hyperliquid's automated market-making pools as market makers for the US venue, or act as a broker into Hyperliquid's markets. It can also adopt Hyperliquid's risk engine and matching engine; Mansi suggests Coinbase would value the technology more than the liquidity. Offering Hyperliquid as-is to US customers doesn't work. Every route needs a regulated intermediary that holds customer assets, owes customers a fiduciary duty, and assesses the risk of each external protocol.
The CFTC will limit its jurisdiction to prediction-market contracts and leave out legacy casinos, Mansi Birla notes. In October it proposed treating certain prediction-market contracts, sports included, as swaps, and it issued an interim rule that excludes casino-style products and licensed sportsbook wagers from the swap definition. Federal appeals courts have split on sports contracts, and New Jersey has asked the Supreme Court to review the question. Mansi doesn't predict the outcome. She says revenue is a big part of the states' motive: gambling has historically been a state matter, and she suggests some gaming incumbents feel the pinch as sports bets move into prediction markets.
Prediction markets carry four types of risk in Mansi Birla's view: participants who influence the event itself, information asymmetry, disputes over how a contract resolves, and manipulation of the published probability. She finds the last one the most interesting. Journalists, analysts, and businesses quote these markets as forecasts. That gives someone a reason to move the displayed odds for reputational or strategic ends, even when trading profit isn't the main goal. In traditional markets a bad actor manipulates the price. In prediction markets, the target can be the event itself or the public reading of the price.
Asked to choose between Kalshi's centralized resolution and Polymarket's decentralized approach, Mansi Birla, who has worked on the Kalshi model, backs Kalshi. Centralized resolution leaves less room for debate, and a regulated intermediary answers to a regulator with clear standards. She sees that as better for retail. A decentralized design offers less scope for manipulation and less exposure to a single point of failure, but Mansi notes that most protocols are not truly decentralized.
Smaller players won't have it easy, Mansi Birla says. DCM applications have surged since the new rules arrived, and no regulator will grant that many licenses. A new venue needs the right talent, a lot of capital, and a partnership with a clearing house, and the few DCOs that exist mostly belong to larger players. Outside the US the map is more fragmented. Europe and the UK ban binary options trading for retail clients, a class that can include many prediction-market contracts, and a firm that passports a derivatives license across the EU still needs gaming licenses country by country. Even in the US, gaming companies are fighting federal jurisdiction over these products.