Speaking Wednesday (Oct. 28) at the agency’s inaugural Frontier Forum in Washington, Selig linked two emerging challenges: developing markets for contracts tied to computational resources and adapting financial oversight to software that can identify opportunities, hold assets and execute transactions for users.
His remarks offered a direction for regulatory development rather than a specific rulemaking proposal. Selig also cautioned that his views did not necessarily reflect those of the commission.
Selig identified computational power, commonly called “compute,” as a resource that could generate demand for new derivatives markets. Advanced AI systems depend on specialized processors, electricity and data center infrastructure. Rising demand creates exposure to changes in both resource prices and availability.
He compared the potential market with established uses of derivatives, such as farmers hedging crop prices, airlines managing fuel costs and manufacturers protecting against fluctuations in raw materials.
AI developers, infrastructure providers, cybersecurity companies and investors could likewise need financial instruments to manage their exposure to the resources powering AI development and deployment.
“Earlier this year, the CFTC sought public comment on the development of compute derivatives markets,” Selig said. More than 100 commenters responded, addressing the opportunities and risks.
For businesses committing capital to AI infrastructure, the potential significance extends beyond another trading product. Compute futures could provide a mechanism for managing uncertain operating costs and informing investment decisions, applying the risk-transfer function of derivatives to an emerging economic resource.
Selig did not provide specifics on contract specifications or announce approvals. Instead, he emphasized the need to understand how these markets are evolving and how the agency’s framework could support their responsible development.
Alongside those emerging potential markets, Selig pointed to a more fundamental regulatory challenge of assigning responsibility when software performs functions traditionally carried out by people and financial intermediaries.
The CFTC’s framework assumes identifiable participants performing defined financial activities. AI agents challenge those assumptions by conducting transactions with increasing independence.
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
“Does the developer of that agent become subject to registration? Who is liable if the agent goes rogue?” Selig asked.
He also questioned how regulators would determine whose interests an agent, or group of agents, represents, raising the possibility of a “know-your-agent” approach alongside existing know-your-customer requirements.
These were questions for further examination, rather than announced registration obligations or a new identity-verification standard, he said.
Selig pointed to the commission’s experience with crypto developers, who have faced uncertainty about whether publishing software or creating interfaces triggers requirements intended for traditional intermediaries.
“Our approach must remain grounded in the functions performed, the control exercised, and the economic realities of the transaction,” he said.
That approach puts the allocation of authority and control at the center of the regulatory project as developers, users and intermediaries rely more heavily on autonomous systems.
Selig acknowledged that agents could react unexpectedly to market conditions. Systems using similar models could produce correlated trading, while rapid autonomous transactions could complicate surveillance and risk management.
He also argued that AI could strengthen oversight by helping exchanges detect suspicious activity, helping clearinghouses monitor exposures and providing regulators a tool to detect misconduct. CFTC staff already use AI to police markets, he said.
The broader opportunity, in his account, is wider access to sophisticated analysis and hedging capabilities previously concentrated among large institutions.
For market participants, the speech signaled two areas to watch. They include the infrastructure supporting compute derivatives and the rules assigning responsibility for agentic trading. Both remain developing policy questions, with Selig advocating engagement between technology builders and regulators as the agency considers how existing oversight should evolve.
Add as Preferred Source
