The U.S. Commodity Futures Trading Commission has formally requested public comment on how exchanges should list and oversee derivatives tied to artificial intelligence computing capacity. The 19-page request, identified as RIN 3038-AF77, was issued on August 19 and focuses on cash market liquidity, reference prices, manipulation controls, customer safeguards and the possible use of perpetual compute futures.
Comments will remain open for 60 days after the document is published in the Federal Register. As of August 20, that publication had not yet occurred, so the final submission deadline had not been set. The request does not propose a rule, approve a contract or authorize trading; it is an initial regulatory information-gathering step.
CFTC Chairman Michael S. Selig said the U.S. “cannot win the AI race without a robust derivatives market for compute.” The agency said the underlying commodity would typically be access to rented computing capacity, such as the hourly rental price of an Nvidia B200 processor, though a contract could also reference a specified quantity of AI inference tokens.
At the center of the inquiry is whether the compute market is liquid and transparent enough for reliable futures settlement. The CFTC noted that much of the economic value in compute markets appears to flow through private bilateral agreements, while prices vary across providers, regions, processor models and contract structures. Regulators are asking whether cash-settled contracts should rely on prices they cannot fully observe, verify or monitor, and they want feedback on surveillance, information sharing and safeguards against distorted settlement prices.
The request also asks whether perpetual compute futures could offer risk-management functions that fixed-expiry contracts cannot. Perpetual contracts have no standard expiry and typically use recurring funding payments to stay aligned with the referenced market. The issue follows the CFTC’s broader examination of perpetual derivatives; the agency has allowed certain crypto perpetual contracts, though CME Group is challenging that policy in federal court.
Silicon Data, a startup focused on financial infrastructure for AI, is developing pricing benchmarks for compute resources. The company is building indices that track the cost of AI compute, aiming to give financial institutions a way to value and hedge GPU and data center costs. CME Group and Silicon Data plan to launch two cash-settled compute futures contracts on October 5, subject to regulatory review. One contract would represent one month of Nvidia H100 rental costs, while the other would track rental costs for Nvidia’s newer Blackwell B200 processor. The planned launch date does not guarantee regulatory review will finish by then.
The CFTC also requested input on retail disclosures, anti-money laundering controls, position limits and the geopolitical sensitivity of advanced computing capacity, encouraging empirical and data-based submissions.
