How does Raymond James govern AI and manage AI risk?
Raymond James treats responsible AI as model governance, data containment and human review, and its filings show how quickly its position changed. The fiscal 2024 10-K, filed November 26, 2024, said 'we currently do not use AI extensively'; the fiscal 2025 10-K a year later says the firm uses, develops and incorporates AI including generative AI and relies on proprietary or 'walled-garden' environments to contain security and privacy risk. The June 2025 investor day listed 'Safety and Responsibility' among five AI guideposts: AI models must undergo 'strict model governance and adherence to current and future regulations'. Model Risk Management is a separate department inside Risk Management, independent of model owners, and reports validation issues to the Enterprise Risk Management Committee and the board's Risk Committee; the 2023 annual report said no AI innovation reaches production until model governance teams have tested and reviewed it. Chief AI officer Stuart Feld said in March 2025 that vendor contracts bar vendors from training on or retaining firm data, and the firm keeps 'human checkpoints' on AI Search and 'full human-in-the-loop oversight' on its Rai agent. On regulation, general counsel Jonathan Santelli urged the SEC in October 2023 to withdraw its predictive data analytics proposal, which the SEC did in June 2025. The 2026 proxy records board education on AI but names no committee as owner of AI, and the firm has published no AI principles document.[1][2][3][4][5][6][7][8][9][10]
Data as of Oct 31, 2026 · Raymond James profile updated Oct 13, 2026 · 20 sources cited on this page · 141 banks profiled on this site · Markdown
Who oversees AI at Raymond James?
'Disclosed' means it is in an SEC filing, the annual report or a firm release; 'Stated' means a named executive said it in an interview.
| Body or role | What it does for AI | Status | Source date |
|---|---|---|---|
| Risk Committee (board) | Receives model validation issues from Model Risk Management, as does the Enterprise Risk Management Committee | Disclosed | November 25, 2025[2] |
| Model Risk Management | A separate department within Risk Management, independent of model owners, users and developers; covers model governance, the firmwide model inventory, validating and approving models, and ongoing monitoring | Disclosed | November 25, 2025[2] |
| Model governance teams | 'No AI innovation moves from the lab into production until it has been thoroughly tested and reviewed by our model governance teams' | Disclosed | December 2023[4] |
| Carillon Labs | Technology team created in 2019 that explores and tests data science and machine learning on back-office problems | Disclosed | December 2023[4] |
| Technology Advisory Council | Experienced advisors, nominated across affiliation models, who guide the technology teams developing AI tools for advisors | Disclosed | December 2023[4] |
| Chief AI officer | Stuart Feld: the mandate is 'how do we leverage AI in a responsible and productive way?'; the firm understands what models do 'by continuously monitoring them' | Stated | March 10, 2025[5] |
| Technology Review Board | Chaired by chief architect Seth Ford from June 2026; it 'guides technology decisions to promote sound architectural practices, manage risk and ensure long-term operational sustainability' | Disclosed | June 8, 2026[11] |
| Full board | Director education sessions have covered 'advances in artificial intelligence and related technologies'; the proxy names no committee as owner of AI | Disclosed | January 7, 2026[10] |
What has Raymond James disclosed about AI risk and governance, and when?
| Date | Where | What it says |
|---|---|---|
| October 9, 2023 | Comment letter to the SEC | General counsel Jonathan Santelli urges the SEC to withdraw its predictive data analytics proposal; its 'covered technology' definition would reach even 'electronic reminders for customers to consider rebalancing'[8][12] |
| November 21, 2023 | Fiscal 2023 10-K | AI appears once, in the privacy discussion: 'technology advances in the areas of artificial intelligence, mobile applications, and remote connectivity solutions have increased the collection and processing of personal information'[13] |
| December 2023 | 2023 annual report | 'Governance is at the forefront of our AI development process, with guide rails in place to assure models align with their intended purpose'; Feld: 'The firm's incremental approach is the right one'[4] |
| November 26, 2024 | Fiscal 2024 10-K | 'Although we currently do not use AI extensively, we may in the future use, develop, and incorporate' AI including generative AI; warns of algorithms 'based on biased, incomplete, and/or inaccurate datasets' and cites the E.U. Artificial Intelligence Act[1] |
| March 10, 2025 | Financial Planning interview, Stuart Feld | 'Security is first and foremost for us'; vendors 'cannot train and even retain information that is in their systems'; the firm uses other companies' large language models and would deal only with established companies such as Microsoft and Google[5] |
| April 17, 2025 | AI Search launch | Feld: 'our practice is to maintain human checkpoints on this and any future systems throughout development, implementation and maintenance'; a real-time voting system rates the answers[6] |
| June 5, 2025 | 2025 investor day | Five-part AI framework ending with 'Safety and Responsibility: Ensure AI models undergo strict model governance and adherence to current and future regulations'[3] |
| June 12, 2025 | SEC withdrawal notice | The SEC withdraws the predictive data analytics proposal, with other proposals, as of June 17, 2025, and 'does not intend to issue final rules' on them[9] |
| November 25, 2025 | Fiscal 2025 10-K | 'We use, develop, and incorporate' AI and machine learning including generative AI; security and privacy risk is mitigated with proprietary or 'walled-garden' environments; some U.S. states' AI laws can require notice and, in some cases, consent for use of AI[2] |
| January 27, 2026 | Rai launch | The operations agent works 'while maintaining full human-in-the-loop oversight' and learns to consider each user's 'role, responsibilities and entitlements'[7] |
| April 22, 2026 | Q2 fiscal 2026 call, Paul Shoukry | A family's financial relationship is 'not something I would trust to an AI bot no matter how good the algorithm is'[14] |
What has Raymond James not disclosed about AI governance?
Raymond James has not published AI principles or a responsible AI policy. Its 2026 proxy records director education on AI but names no board committee as owner of AI oversight, and the 10-K describes the model inventory without saying how many models use AI. The three 10-Qs for fiscal 2026 carry no AI language; the AI risk factor sits in the annual 10-K.[10][2][15]
The firm's main recorded intervention on AI rules is the October 2023 letter, in which Santelli argued that existing fiduciary and Regulation Best Interest standards already govern advice and that conflicts can be addressed by disclosure. The letter notes that Raymond James is a bank holding company supervised by the Federal Reserve; no AI-specific supervisory finding against it appears in its filings.[8][2]
Feld has also set limits on use: in March 2025 he said AI-generated financial plans are 'not a space where we want to go' for now because the firm needs to 'know exactly what the black box is doing', and of the technology: 'It's not making decisions'.[5]
What does the Raymond James profile record on this topic?
The rows below are the entries of the Raymond James AI strategy profile that bear on this page, as of Oct 13, 2026; the profile holds the full record of 31 dated moves, 16 use cases and 46 sources.
| Use case | What the record shows | Status | Source |
|---|---|---|---|
| AI risk-factor disclosure and walled-garden controls | The 10-K names inaccurate or biased AI output, vendor practices, security and privacy risk and AI regulation, mitigated by proprietary 'walled-garden' environments. | In production | [2] |
| Machine learning in threat detection and fraud prevention | AI enhances real-time threat detection and fraud prevention; on the April 2026 call Shoukry said the firm uses AI in cybersecurity while the tooling continues to evolve. | In production | [16][17] |
| Model governance gate before production | Per the firm, no AI innovation moves from Carillon Labs into production until model governance teams have tested and reviewed it; Seth Ford now chairs a Technology Review Board that guides technology decisions and manages risk. | In production | [16][11] |
| Date | Move | Detail | Source |
|---|---|---|---|
| Aug 5, 2026 | June-quarter 10-Q filed with no reference to AI | The three fiscal 2026 10-Qs carry no AI language; the AI risk factor sits in the fiscal 2025 10-K filed on 25 November 2025. | [15][18][19][2] |
| Jun 8, 2026 | Seth Ford hired as chief architect to lead AI-ready architecture | Formerly Chief Architect and Head of Agentic Digital Products at Vanguard; he chairs the Technology Review Board and does not report to Feld, though both sit under the CIO's organisation. | [11][20] |
| Jan 7, 2026 | Proxy statement records board education on AI | The 2026 DEF 14A says director education sessions have covered advances in artificial intelligence alongside data privacy and cybersecurity. | [10] |
| Nov 25, 2025 | 10-K: generative AI in the platform, 'walled-garden' controls | The fiscal 2025 annual report says the firm uses, develops and incorporates AI and machine learning including generative AI, and flags inaccurate or biased output, security and privacy risk, and AI regulation costs. | [2] |
| Apr 17, 2025 | AI Search launched in-house under chief AI officer Feld (appointed February) | Generative-AI search over the internal knowledge base, with a real-time voting system on answer quality and human checkpoints throughout development and maintenance. | [6] |
| Mar 10, 2025 | CAIO interview: no in-house LLMs, vendors barred from training on firm data | Feld says the firm uses other providers' models, would deal only with established companies such as Microsoft and Google, has contracts under which vendors cannot train on or retain its information, and is not yet using generative AI for financial planning. | [5] |
| Aug 1, 2024 | Firm describes Carillon Labs and its model governance gate | Carillon Labs, created in 2019, tests machine learning for back-office use; algorithms cut false positives in electronic-communications review, AI supports cybersecurity, fraud prevention and AML due diligence, and no AI project reaches production until model governance teams have tested it. | [16] |
Does Raymond James have responsible AI principles?
Not as a published document. The closest statement is the 'Safety and Responsibility' guidepost in its June 2025 investor day framework, which commits to 'strict model governance and adherence to current and future regulations', plus the 2023 annual report's rule that no AI moves from the lab to production before model governance teams test and review it.
Who oversees AI risk at Raymond James?
Model Risk Management, a department within Risk Management that is independent of model owners and developers, validates and monitors models and reports issues to the Enterprise Risk Management Committee and the board's Risk Committee. Chief AI officer Stuart Feld leads AI work, and chief architect Seth Ford chairs the Technology Review Board. The 2026 proxy names no board committee for AI.
Does Raymond James let AI vendors train on client data?
No, according to chief AI officer Stuart Feld. In March 2025 he said vendor contracts stop vendors from training on or retaining information in their systems, that none of the firm's information is in open, trainable models and that data is encrypted. The fiscal 2025 10-K says the firm uses proprietary or 'walled-garden' environments to limit AI security and privacy risk.
What did Raymond James tell the SEC about AI rules?
In an October 9, 2023 letter, general counsel Jonathan Santelli urged the SEC to withdraw its proposed rule on conflicts of interest in predictive data analytics, arguing its 'covered technology' definition would reach even spreadsheets and rebalancing reminders. The SEC withdrew the proposal in a notice dated June 12, 2025, effective June 17, 2025.
Does Raymond James use AI to give investment advice?
Not directly, by its own account. Feld said in March 2025 that AI-generated financial plans are not a space the firm wants to enter yet, and that the technology is not making decisions or replacing advisors. CEO Paul Shoukry said in April 2026 he would not trust his family's financial relationship to an AI bot. Its AI tools serve advisors and associates, with human review.
- Raymond James Financial Annual Report on Form 10-K for the fiscal year ended September 30, 2024
- Raymond James Financial Annual Report on Form 10-K for the fiscal year ended September 30, 2025
- Raymond James Financial 2025 Analyst & Investor Day presentation (Form 8-K, Exhibit 99.1)
- Raymond James Financial 2023 Annual Report (Prudently approaching AI possibilities)
- Raymond James' new AI chief opens up about tech at the firm
- Raymond James Invests in Service Excellence with Proprietary Generative AI Search
- Raymond James launches proprietary digital AI agent to bolster operational excellence
- Raymond James Financial comment letter on Conflicts of Interest Associated with the Use of Predictive Data Analytics (File No. S7-12-23)
- Withdrawal of Proposed Regulatory Actions (Release No. 33-11377)
- Raymond James Financial 2026 Proxy Statement (DEF 14A)
- Raymond James appoints Chief Architect to bolster enterprise technology and AI strategy
- Advisors contend SEC crackdown on AI recommendations does not compute
- Raymond James Financial Annual Report on Form 10-K for the fiscal year ended September 30, 2023
- From focus groups to coaches: This is how firms are handling the AI revolution
- Raymond James Financial Quarterly Report on Form 10-Q for the quarter ended June 30, 2026
- Prudently approaching AI possibilities
- Earnings call transcript: Raymond James beats Q2 2026 forecasts amid tech investments
- Raymond James Financial Quarterly Report on Form 10-Q for the quarter ended March 31, 2026
- Raymond James Financial Quarterly Report on Form 10-Q for the quarter ended December 31, 2025
- Raymond James taps Vanguard's chief architect to drive AI strategy
When Raymond James moves on AI, you'll read it here first.
when one of the banks we cover moves on AI, it is in the next morning's brief · six sourced stories · 7 am PT · free
plus every regulation, bank and agent page update, the morning after · leave any morning