On December 5, 2024, the Commodity Futures Trading Commission (“CFTC”) issued an inter-divisional staff advisory (the “Advisory”) on the use of artificial intelligence (“AI”) by CFTC-regulated entities, including market infrastructures and intermediaries, in CFTC-regulated markets.[1] The Advisory reinforces the CFTC’s technology-neutral stance while reminding CFTC-regulated entities to comply with the Commodity Exchange Act (“CEA”) and CFTC regulations. The key message from the CFTC is that a regulated entity must comply with all applicable CEA and CFTC requirements in its adoption and integration of AI, regardless of whether it develops the AI technology internally or procures it from a third-party service provider.[2]
Staff provided a few examples of potential changes a CFTC-regulated entity may need to make to its compliance program. Specifically, such an entity should “assess the risks of using AI and update policies, procedures, controls, and systems, as appropriate.”[3] In addition, to the extent the integration of AI results in a material system or process change, the Advisory reminds regulated entities to “ensure that [the] adoption of AI has been reviewed for compliance with the CEA and CFTC regulations,” including any regulatory obligation to “give Staff timely advance notice of all material planned changes to automated systems that may impact the reliability, security, or adequate scalable capacity of such systems.”[4] The Advisory notes that “Staff may also incorporate AI as a topic of discussion in its routine oversight activities such as examinations” and therefore may feature prominently as an examination priority or the subject of investigations or enforcement actions.[5] Notably, Commissioner Kristin Johnson, in her supporting statement, reiterates her call for the CFTC to “enhance the safety and benefits of incorporating AI in markets while minimizing the risks,” and advocates for the creation of an AI Fraud Task Force.[6]
This Advisory reflects the CFTC’s efforts to clarify its perspective on AI use ahead of the presumptive rapid adoption to come in the near future, particularly of generative AI and large language models, and sets minimum standards for how AI technology should integrate within derivatives markets. The Advisory draws from many sources, including the AI Task Force Chairman Rostin Behnam assembled, materials from the Market Risk Advisory Committee, and public comments submitted in response to the CFTC’s Request for Comment on the Use of AI in CFTC-Regulated Markets.[7] The CFTC acknowledges that it must continue to evaluate the need for future regulation, guidance, or other CFTC or Staff action as novel technology is integrated into various industry processes.
Below is a table summarizing how the CFTC anticipates the various entities subject to its jurisdiction may implement AI technology and the CEA and CFTC provisions that might be implicated.
Summary Table
The Advisory applies to designated contract markets (DCMs), swap execution facilities (SEFs), swap data repositories (SDRs), derivatives clearing organizations (DCOs), futures commission merchants (FCMs), swap dealers (SDs), commodity pool operators (CPOs), commodity trading advisors (CTAs), introducing brokers (IBs), retail foreign exchange dealers (RFEDs), and Associated Persons.
| Use of AI in CFTC-Regulated Markets | Applicable CEA and CFTC Regulations |
| DCMs, SEFs, SDRs | |
| Order Processing and Trade Matching. DCMs may use AI in their order processing and trade matching functions. For example, DCMs could use AI to anticipate trades before they happen to proactively allocate system resources to reduce post-trade message latencies.[1] | Core Principle 9 (Execution of Transactions). |
| Market Surveillance. DCMs and SEFs may use AI to detect abusive trading practices, investigate rule violations, monitor markets in real time, or engage in other regulatory functions.[2] For example, a DCM or SEF could use AI to identify “frontrunning, wash trading, pre-arranged trading, fraudulent trading, money passes,” as well as trade execution patterns and anomalies. | DCMs: Core Principle 2 (Compliance with Rules); Core Principle 4 (Prevention of Market Disruption); and Core Principle 12 (Protection of Markets and Market Participants).
SEFs: Core Principle 2 (Compliance with Rules); Core Principle 3 (Swaps Not Readily Susceptible to Manipulation); and Core Principle 4 (Monitoring of Trading and Trade Processing). |
| System Safeguards. The Advisory does not provide specific predictions of how DCMs, SEFs, and SDRs may use AI in their system safeguards program other than acknowledging that they may use AI-based products from third-party service providers.[3] | DCMs: Core Principle 20 (System Safeguards); CFTC Regulation 38.1051.
SEFs: Core Principle 14 (System Safeguards); CFTC Regulation 37.1401. SDRs: CFTC Regulation 49.24 (System Safeguards). |
| DCOs | |
| System Safeguards. DCOs may use AI to identify cyber vulnerabilities and intrusions and improve cyber defenses or evaluate and update legacy and new computer code.[4] | Core Principle B (Financial Resources); Core Principle D (Risk Management); Core Principle I (System Safeguards); and CFTC Regulation 39.18(d)(2). |
| Use of AI in CFTC-Regulated Markets | Applicable CEA and CFTC Regulations |
| Member Assessment and Interaction. DCOs may use AI to review whether their clearing members are complying with DCO rules or to communicate with the members, including through chatbots.[5] This may become more prominent in non-intermediated clearing. | Core Principle C (Participant and Product Eligibility) and Core Principle D (Risk Management). |
| Settlement Procedures. DCOs may use AI to support settlement or facilitate netting or offset positions by validating data, mining data anomalies prior to settlement, or identifying failed trends.[6] | Core Principle E (Settlement Procedures). |
| FCMs, SDs, CPOs, CTAs, IBs, RFEDs and Associated Persons | |
| Risk Assessment and Management. Registrants may use AI to calculate and collect initial and variation margin for uncleared swaps under CFTC Regulation 23.152.[7] | Staff did not identify a specific rule or regulation that could apply to this use case. |
| Compliance and Recordkeeping. Registrants may use AI to support the accuracy and timeliness of financial information and risk disclosures submitted to the CFTC, the National Futures Association, and customers.[8] For example, a CPO may use AI to update a disclosure document or prepare account statements for a commodity pool. | Part 4 of the CFTC’s regulations and any other rules or regulations related to financial information and risk disclosures. |
| Customer Protection. Registrants may use AI to account for segregated customer funds.[9] | Part 1 of the CFTC’s regulations. |
Click here to download this article.
[1] Advisory at 3.
[2] Id.
[3] Id.
[4] Id. at 4.
[5] Id.
[6] Id. at 4-5.
[7] Id. at 5.
[8] Id.
[9] Id.
[1] CFTC Release No. 9013-24, CFTC Staff Issues Advisory Related to the Use of Artificial Intelligence by CFTC-Registered Entities and Registrants (Dec. 5, 2024), available here.
[2] Id. at 2.
[3] Id.
[4] Id. at 2-3.
[5] Id. at 5.
[6] Statement of Commissioner Kristin N. Johnson on Future-Proofing Financial Markets: Assessing the Integration of Artificial Intelligence in Global Derivatives Markets (Dec. 5, 2024), available here.
[7] Advisory at 2; see Market Risk Advisory Committee, Transcript of Meeting on Apr. 9, 2024, available here. Tamika Bent, as Chief Counsel to Commissioner Kristin Johnson, was the Designated Federal Officer of the Market Risk Advisory Committee during this meeting.
[8] Advisory at 3.
[9] Id.
[10] Id.
[11] Id. at 4.
[12] Id.
[13] Id. at 4-5.
[14] Id. at 5.
[15] Id.
[16] Id.
