International Council for Derivative Trading (ICFDT) Comment On Regulatory Notice 26-14 | FINRA.org Skip to main content

International Council for Derivative Trading (ICFDT) Comment On Regulatory Notice 26-14

International Council for Derivative Trading (ICFDT)

INTERNATIONAL COUNCIL FOR DERIVATIVE TRADING

Issuer of the CFOA and CFIA professional designations

August 18, 2026

Jennifer Piorko Mitchell

Office of the Corporate Secretary

Financial Industry Regulatory Authority

1700 K Street NW

Washington, DC 20006

Re: Regulatory Notice 26-14, Proposed Changes to Modernize Rule 2210 (Communications with the Public)

Dear Ms. Mitchell,

The International Council for Derivative Trading (ICFDT) submits this comment in response to Regulatory Notice 26-14.

ICFDT is a credentialing body for professionals working in derivative markets. It issues the Certified Futures and Options Analyst (CFOA), a certification in futures, options, volatility, and risk management, and the Chartered Financial Intelligence Architect (CFIA), a designation for investment professionals who govern and oversee the use of artificial intelligence in investment workflows. Each is awarded on the basis of a proctored examination against a published body of knowledge.[1] A CFOA certificate is issued through an independent third party credentialing platform and carries a unique verification link, so a firm can confirm a holder's certification at the time of review without relying on the holder's description of it, and a certification is subject to a published professional conduct procedure under which it may be withdrawn.[2] The CFIA additionally maintains a public charterholder registry, continuing professional development requirements, and a code of professional conduct enforced through a conduct panel. ICFDT is not a FINRA member and does not act as one.

ICFDT confines this comment to one element of the proposal: the treatment of preparer qualifications as a risk factor under proposed Supplementary Material .01. This bears on Requests for Comment 1(b), 2(b), and 3(c). ICFDT takes no position on the filing requirement changes, on the proposed amendments to Rule 2210(d)(7), or on the economic impact assessment.

ICFDT discloses its interest at the outset. It issues credentials. A rule that assigns evidentiary weight to credentials would benefit it. The argument below is therefore framed so that it does not depend on any particular credential, including ICFDT's own, and the recommendation at Part V names no designation and asks FINRA to endorse none.

Summary of Position

1     ICFDT supports replacing mandatory principal pre-use approval with a risk-based supervisory standard.

2     The second factor in proposed Supplementary Material .01, the qualifications and experience of the preparer, is currently unoperationalized. A risk factor that cannot be evidenced at the time of review cannot be defended in an examination and will in practice be dropped from firms' written procedures.

3     The distinction that gives the factor content is between a qualification claim that is confirmable and revocable by a party other than the claimant, and one that is not. FINRA can draw that distinction without approving, endorsing, or accrediting any credential, and it should not do any of those things.

4     The factor bites hardest on artificial intelligence generated communications, where authorship is decoupled from review. FINRA should confirm that the qualification factor attaches to the reviewer of record.

I. The Risk-Based Standard is the Right Direction

The current requirement that a registered principal approve each retail communication prior to use was designed for a communications environment in which a firm produced a countable number of discrete pieces. That environment no longer exists. The Notice records that FINRA's Advertising Regulation Department reviewed 172,898 retail communications filed after first use over three years, and that firms are now generating communications at machine speed. Uniform pre-use approval applied across that volume allocates the same supervisory attention to a scheduled market commentary and to a solicitation for a complex product. ICFDT supports the proposed change.

The proposal's durability depends on the factor list in Supplementary Material .01 being workable. A risk-based system is only as defensible as the inputs it grades on.

II. Factor Two Cannot Presently Be Applied

The proposal directs members to consider:

the qualifications and experience of the preparer of the communication, including persons paid for or involved in the preparation of the content or who explicitly or implicitly endorse or approve the content.

Every other factor in the list can be determined by inspection at the moment of review. Product complexity is knowable from the product. Whether a communication makes a recommendation is knowable from the text. Medium, distribution method, audience tailoring, and the presence of performance data are all observable in the artifact itself. Prior review history is retrievable from the firm's own records.

Factor two is different. It requires a judgment about a person, and the proposal supplies no standard for reaching it. A supervisor asked to grade the qualifications of a paid contributor, an outside content producer, or a compensated social media promoter has three sources: the person's registration status, whatever the person says about themselves, and the supervisor's impression. The first is already fully evidenced through CRD and is therefore not the case the factor is aimed at. The second and third are not evidence.

The predictable consequence is that firms will write factor two into their procedures in the language of the rule and will not apply it, because there is no way to document its application to an examiner. A factor that survives in a procedures manual and dies in practice is worse than an absent factor, because it creates the appearance of coverage.

III. The Finfluencer Case Illustrates the Problem

The Notice states that communications posted by financial influencers may present risks due to the influencer's potential lack of qualifications or conflicts of interest, and that these risks persist regardless of whether content is static or interactive. ICFDT agrees, and observes that the two named risks are not symmetrically tractable.

Conflicts of interest are already addressable. Compensation arrangements are documentable, disclosure obligations attach to them, and a firm can evidence what it paid and what was disclosed. Qualifications are not addressable under the proposal as drafted, because the proposal offers no way to distinguish a compensated promoter who holds an independently confirmable qualification from one who describes themselves in identical terms and holds nothing. Both present the same artifact to the supervisor: a claim.

The distinction that matters is not the prestige of a credential, the size of its issuing body, or the difficulty of its examination. It is whether a party other than the claimant can confirm the claim and withdraw it. A qualification that can be confirmed by a third party at the time of review, and that can be revoked by that third party for conduct, is evidence. A qualification that cannot be confirmed is a statement by the person being assessed about the person being assessed.

IV. Application to Artificial Intelligence Generated Communications

The Notice observes that members may consider the qualifications and experience of the individuals responsible for supervising, validating, or reviewing an artificial intelligence tool. This is where factor two carries the most weight and where its present indeterminacy is most consequential.

In a conventional communication, authorship and accountability sit with the same person, and the supervisor is grading a known individual's work product. Where content is machine generated at volume, authorship is distributed across a model, a prompt, a template, and a validation step, and the only human whose qualifications are meaningfully assessable is the person who signed off. If the rule does not say so, firms will apply factor two to whoever pressed the button.

FINRA should state in guidance that where a retail communication is generated in whole or substantial part by an automated tool, factor two is assessed with respect to the individual who performed or supervised the validation step, and that the identity of that individual should be recorded. This imposes no new obligation. It clarifies where an existing one lands.

V. Recommended Guidance Language

FINRA can give factor two content by identifying evidentiary characteristics rather than by naming credentials. The following language, or language to the same effect, could be added to Supplementary Material .01 or issued as guidance:

In considering the qualifications and experience of the preparer, a member may give greater weight to a qualification that can be confirmed at the time of review by a source independent of the preparer, and that is subject to withdrawal by that source. Registration status reflected in the Central Registration Depository is one example. A credential is another, where the issuing organization provides a means by which a third party can confirm the credential at the time of review and maintains a published disciplinary process under which the credential may be withdrawn. A qualification asserted by the preparer and not independently confirmable should not, without more, reduce the level of supervisory review applied to a communication.

ICFDT emphasizes what this language does not do. It does not approve, endorse, or accredit any designation. FINRA states in its Professional Designations database that it does not approve or endorse any professional credential, and ICFDT agrees that FINRA should not begin. The language is evidentiary. It tells a firm what kind of thing counts as documentation, and leaves to the firm the question of whether any given qualification is relevant to the communication at hand. It assigns no weight to an issuer's reputation, marketing, or age, and none to the prestige or difficulty of a credential.

VI. Recommendations

1     Adopt the risk-based supervisory standard as proposed.

2     Give factor two content along the evidentiary lines set out in Part V, distinguishing independently confirmable and revocable qualifications from asserted ones.

3     Confirm in guidance that for machine-generated retail communications, factor two attaches to the individual who performed or supervised validation, and that this individual is identified in the member's records.

4     Decline to approve, endorse, or accredit any professional designation, and state so expressly in any guidance issued under factor two.

ICFDT appreciates the opportunity to comment and is available to FINRA staff on any point raised here.

Respectfully submitted,

International Council for Derivative Trading

icfdt.com

[email protected]

[1] The CFOA body of knowledge is published at https://icfdt.com/the-cfoa-program/. The CFIA body of knowledge is published at https://icfdt.com/cfia/.

[2] The CFOA Professional Conduct procedure is published at https://icfdt.com/cfoa-conduct/.