Kalshi Builds a Forward Curve for Computing Power as Exchanges Race to Turn GPUs into a Tradable Commodity
Kalshi, the prediction markets exchange, has developed a forward curve that tracks future GPU rental prices, joining CME and ICE in financializing AI infrastructure. This tool utilizes weekly and monthly event contracts related to compute prices, extending up to a year into the future. An algorithm then combines these contracts to create a single curve serving as a reference for futures, options, and other derivatives.
"We are using prediction markets to build the forward curve, which will provide the market a view of what compute costs will be in the future for different grades and time-frames of GPUs," said Udesh Jha, Kalshi’s chief risk officer, according to Bloomberg.
Forward curves, commonly used in commodity markets to predict future prices of various goods, now include GPU rental costs, reflecting the growing trend of treating compute as a commodity.
Kalshi’s approach differs from its competitors in that it leverages existing prediction market frameworks using event contracts already trading, rather than creating traditional futures contracts requiring regulatory approval. This enables hedging, risk management, and speculation alike.
The drive behind this development is the projected trillion-dollar AI infrastructure spending within the next decade. Without a standardized way to hedge against price fluctuations, GPU rental rates have been volatile, and the market for compute remains fragmented across cloud providers, data center operators, and GPU brokers. A functioning forward curve provides a shared view of future prices, forming the basis for hedging and risk management strategies. Whether Kalshi, CME, or ICE ultimately captures the most liquidity will determine which curve becomes the industry standard, much as competing oil contracts have done in the past.