Crypto-style perpetual futures and prediction markets have reached AI computing capacity ahead of planned regulated futures products from CME Group and Intercontinental Exchange (ICE), according to a research note from Bernstein. The development signals growing financialization of computing infrastructure as demand for AI processing power surges.
Perpetual swaps, a type of derivatives contract popularized in cryptocurrency markets, are now being applied to AI compute capacity. These instruments allow buyers and sellers to speculate on or hedge against the price of GPU computing power without taking physical delivery of hardware. The contracts use a funding rate mechanism similar to crypto perpetuals to maintain price alignment with underlying spot markets.
Bernstein noted that the emergence of compute derivatives reflects the maturation of AI infrastructure as an asset class. With demand for GPUs and other AI accelerators outstripping supply, market participants are seeking ways to manage price risk and secure future capacity. Crypto-native trading mechanisms offer a template for how these markets can develop.
CME and ICE are both working on regulated futures contracts tied to AI computing capacity, with Bernstein estimating potential launch dates in late 2026. These institutional-grade products would provide greater transparency and oversight compared to the currently unregulated perpetual swap market. The CME has already built significant expertise in cryptocurrency derivatives through its Bitcoin and Ether futures products.
The convergence of crypto trading infrastructure with AI computing markets represents a new frontier for both industries. As AI compute becomes a traded commodity, the pricing mechanisms, custody solutions, and settlement systems developed for digital assets may find broader application in the computing sector.
This article was adapted from The Block. Read the original here.
