Bernstein said in a new research report, cited by The Block, that the perpetual futures structure developed in crypto has already made its way into the AI compute market ahead of more traditional products. The report points to Architect AX, a Bermuda-registered exchange outside CFTC oversight, which has listed GPU compute perpetual contracts that use a funding-rate mechanism to track spot pricing. By contrast, cash-settled compute futures from CME Group and Intercontinental Exchange are still under CFTC review and are expected by the end of 2026. Bernstein also noted that Kalshi, a CFTC-regulated prediction market platform, had already launched contracts tied to GPU rental prices and on July 14 introduced pricing curves for B200, H200, and A100 chips. In that set, the B200 was quoted at $5.41, versus a historical peak of $7.39. The firm said unused GPU compute cannot be stored, making hedging the only way to manage price risk, while speculative liquidity and the difficulty of building benchmark indices remain major constraints.
Bernstein said in its latest research report, cited by The Block, that the perpetual contract structure born in crypto markets has already moved into the AI compute market.
The report said Architect AX, a Bermuda-registered exchange that is not regulated by the U.S. Commodity Futures Trading Commission, has listed GPU compute perpetual contracts. Those contracts anchor to spot prices through a funding-rate mechanism, a structure taken directly from crypto markets.
Meanwhile, cash-settled compute futures from CME Group and Intercontinental Exchange are still awaiting CFTC review and are expected to launch by the end of 2026.
Bernstein also said Kalshi, a CFTC-regulated prediction market platform, had previously listed prediction contracts linked to GPU rental prices. On July 14, Kalshi introduced compute pricing curves for B200, H200, and A100 chips. The B200 was quoted at $5.41, compared with a historical peak of $7.39.
Bernstein added that unused GPU compute cannot be stored, leaving hedging as the only way to take on price risk. Even so, the market is still dominated by speculative liquidity, and the difficulty of building benchmark indices for compute remains a major bottleneck for further development.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
