Back to the Future: CFTC Emphasizes Existing Regulatory Standards for New Al Technologies

Introduction

Advancements in artificial intelligence (AI) continue to reshape the financial services industry, with the growing popularity of generative Al (GenAl) tools prompting increased attention from US regulators. At the end of 2024, the US Commodity Futures Trading Commission (CFTC) issued a staff advisory discussing how the agency believes the larger universe of Al fits within existing regulatory frameworks (CFTC Al Advisory). This article examines key points from the CFTC Al Advisory and compares them to other recent CFTC actions and statements.

Key CFTC Discussions of AI Prior to the Staff Advisory

CFTC Request for Comment

In January 2024, the CFTC issued a request for comment on the use of Al in CFTC-regulated markets (AI RFC). This initiative sought feedback on several critical issues, including whether Al could impede the enforcement of anti-fraud and market manipulation regulations, how entities are modifying governance structures to accommodate AI, and the potential risks of bias in Al algorithms. The AI RFC marked an important first step in gathering data and gauging industry sentiment on the evolving role of Al in financial markets.

Commissioner Johnson’s Remarks

Building on the AI RFC, CFTC Commissioner Kristin Johnson gave a speech to the New York City Bar Association in February 2024, where she proposed three key initiatives:

1. Conducting a comprehensive survey of Al use across market participants.
2. Introducing heightened penalties for bad actors leveraging Al to commit fraud or market manipulation.
3. Establishing an interagency initiative to harmonize Al regulation across financial markets.

Commissioner Johnson's proposals emphasized both the promise and perils of AI, urging the CFTC and other regulators to adopt proactive measures to help ensure market integrity while fostering innovation. In a subsequent statement in December 2024 released around the same time as the CFTC AI Advisory, Commissioner Johnson highlighted the importance of future-proofing financial markets against AI-related risks, calling for regulatory frameworks that are both adaptive and robust.

CFTC Technology Advisory Committee Recommendations to the CFTC on Responsible AI in Financial Markets

The CFTC Technology Advisory Committee (TAC) released a Report on Responsible AI in Financial Markets in May 2024 in which it recommended industry engagement and strengthened guardrails to stem eroding public trust in financial markets, services, and products. The TAC report highlighted a lack of transparency (described as "black box" characteristics) of certain Al decision-making models, output risks posed by training data relied on by Al systems, fairness concerns from Al system biases, and potential mishandling of sensitive data.

Additionally, the TAC advised the CFTC to develop a responsible Al framework to enable Al systems that are safe, trustworthy, fair, robust, transparent, explainable, and private. The TAC also advocated governance structures, citing to the Office of Science and Technology Policy's "Blueprint for an Al Bill of Rights" and the National Institute of Safe Technology (NIST) Al Risk Management Framework.

The TAC further recommended that the CFTC:

1. Host a public roundtable;
2. Define and adopt an Al Risk Management Framework aligned with the NIST AI framwork;
3. Create an inventory of existing regulations related to Al in the sector;
4. Use the regulation inventory to develop a risk assessment and gap analysis addressing Al systems; and
5. Use the gap analysis to guide future CFTC guidance and rulemaking.

Key Points from the CFTC AI Advisory

Rather than focusing on the broader policy initiatives outlined by Commissioner Johnson, the CFTC Al Advisory provides a non-exhaustive list of existing statutory and regulatory requirements that may be triggered by Al use for CFTC-regulated entities. Compliance, depending on the nature of the entity and its operations, may require a thorough assessment of AI-related risks and corresponding updates to registrants' policies, procedures, controls, and systems as appropriate. CFTC-regulated entities must also comply with existing statutory and regulatory requirements without regard to whether they deploy Al internally or through a third-party service provider.

• Designated Contract Markets (DCMs), Swap Execution Facilities (SEFs) & Swap Data Repositories (SDRs). In general, DCMs act as centralized exchanges for futures and options, while SEFs and SDRs act as trading platforms and data repositories for swap transactions, respectively. With respect to order processing and trade matching, DCMs must follow CFTC Core Principle 9, which requires providing competitive, open, and efficient markets and mechanisms for executing transactions that protect price discovery processes, regardless of Al use. DCMs and SEFs also play a key role in market surveillance. The CFTC Al Advisory acknowledges that leveraging Al to detect and flag abusive trading practices, conduct investigations, and monitor markets in real time may be beneficial. The market infrastructure section concludes that DCMs, SEFs, and SDRs must develop and maintain appropriate controls addressing applicable risk categories and provide timely advance notice of all material planned changes to automated systems that may impact reliability, security, or scalability.
• Derivatives Clearing Organizations (DCOs). DCOs generally act as central counterparties for derivatives transactions, managing clearing, settlement, margin requirements, and default risks. The CFTC Al Advisory stresses that DCOs may only leverage internal or third-party Al solutions in compliance with the Commodity Exchange Act of 1936, as amended (CEA) and CFTC regulations. Some DCOs may use Al as a tool to evaluate and/or update legacy and new computer code. The CFTC Al Advisory cautions that the introduction of Al could impact the functionality, reliability, and resilience of computer systems DCOs rely on for their core functions. DCOs may also use Al to review clearing members' compliance with DCO rules or support communications with members through Al chatbots. These reviews and interactions remain subject to Core Principle C (participant admission and eligibility), and Core Principle D (monitoring of credit exposure). DCOs may also use Al to facilitate netting or offset positions as the Al works to validate data, mine data anomalies prior to settlement, or identify failed trades. These activities remain subject to existing laws and regulations on settlement timing, exposure limitations, and netting or offset requirements.
• Introducing Brokers (IBs), Futures Commission Merchants (FCMs), Commodity Pool Operators (CPOs), Commodity Trading Advisors (CTAs), Swap Dealers (SDs) & Retail Foreign Exchange Dealers (RFEDs). The CFTC Al Advisory discusses all these entities and their associated persons in a single section, but each plays a specific role in US markets for futures, options, swaps, and other derivatives (excluding SEC-regulated derivatives, "Commodity Interests") that may impact its use of Al technology. In general: (i) CPOs manage pooled investment vehicles that trade Commodity Interests; (ii) CTAs advise on or manage clients' Commodity Interest trading; (iii) lBs solicit or accept orders for Commodity Interests from their customers; (iv) FCMs receive money and other property from customers for Commodity Interest trading; (v) SDs trade swaps on a principal basis; and (vi) RFEDs receive money and other property from customers for foreign exchange trading.

The CFTC Al Advisory notes that an SD using an Al system to calculate and collect initial and variation margins for uncleared swaps must ensure that the system effectively manages the risk created by such transactions. Any entity above that uses Al in connection with its compliance or recordkeeping operations (e.g., a CPO using a GenAI tool to update a disclosure document or prepare periodic account statements for a commodity pool), or to ensure the satisfaction of its customer protection obligations, remains responsible for ensuring compliance with the CEA and applicable CFTC regulations.

Commissioner Johnson’s Unaddressed Proposals

While the CFTC Al Advisory provides clarity on certain points, it does not address the broader measures advocated by Commissioner Johnson. For example, her proposals for heightened civil monetary penalties, enhanced enforcement resources, and interagency collaboration are not discussed. It is also unclear whether the incoming CFTC leadership will prioritize the same issues, especially given the Trump Administration's recent actions on AI, which included rescinding former President Joe Biden's October 2023 Al Executive Order, ordering the revision of AI-related OMB Memoranda M-24-10 and M-24-18 on which certain CFTC plans were based, and other actions further detailed in another Mintz report.

Conclusion

The CFTC AI Advisory emphasizes that existing regulatory frameworks already govern AI use in futures markets, reinforcing the agency’s commitment to market integrity while allowing innovation to continue. However, it does not address broader policy reforms proposed by Commissioner Johnson and others, leaving open questions about future regulation. As AI evolves, market participants must ensure compliance with current obligations while adapting to emerging risks and regulatory refinements. Proactively assessing AI-related policies, procedures, and risk management frameworks, along with maintaining an open dialogue with regulators, will be crucial for regulated entities to balance AI-driven efficiencies with applicable regulations.

Subscribe To Viewpoints