Independent model review and board oversight, but little disclosure to those affected
Overall rating “solid”. No deduction for controversies. The superscript numbers link directly to the source.
1 · Responsibility
JPMorgan documents board oversight and independent model review, but AI principles, AI figures and complaints channels are missing, and a lawsuit concerns purchased AI.
What is documented
- According to the 2026 proxy, the full board oversees cyber risks and AI matters, supported by the Audit and Risk Committees1
- AI is part of the skills matrix and of board training in 20251 2
- A Chief Data & Analytics Officer on the Operating Committee is responsible for data governance and AI adoption3
- The independent Model Risk Governance and Review function reviews, approves and tiers ML and AI models before deployment4
The case against
Missing rules and figures
Sources for Responsibility (14)
- JPMorgan Chase & Co. (SEC EDGAR): Proxy Statement 2026 (Form DEF 14A) · 3 November 2026 · primary source
- JPMorgan Chase & Co. (SEC EDGAR): Proxy Statement 2025 (Form DEF 14A) · 3 November 2026 · primary source
- JPMorgan Chase & Co.: Teresa Heitsenrether – Leadership · 3 November 2026 · secondary source
- J.P. Morgan Europe Limited (subsidiary, investor relations filing of JPMorgan Chase & Co.): J.P. Morgan Europe Limited, Annual Report for the year ended 31 December 2025 · 9 April 2026 · primary source
- JPMorgan Chase & Co.: AI and model risk governance · 29 May 2023 · secondary source
- American Banker: JPMorgan is working with U.S. regulators on generative AI pilot projects · 9 November 2023 · secondary source
- JPMorgan Chase & Co. (SEC EDGAR): Form 10-K for fiscal year 2025 (jpm-20251231.htm) · 13 February 2026 · primary source
- JPMorgan Chase & Co. (Investor-Relations-Ablage): Form 10-K 2025, PDF version · 13 February 2026 · primary source
- JPMorgan Chase & Co.: 2025 Sustainability Report · 3 November 2026 · secondary source
- Evident Insights: Here's the 2025 Evident AI Index · 3 November 2026 · secondary source
- JPMorgan Chase & Co.: Jamie Dimon, Letter to Shareholders (Annual Report 2025) · 6 April 2026 · secondary source
- US Court of Appeals for the First Circuit (filed via Courthouse News): Mozart Saint Cyr v. JPMorgan Chase & Co., Brief of Appellant, No. 25-2129 · 2 February 2026 · primary source
- Bloomberg Law: JPMorgan Scores Dismissal of Job Seeker's AI Lie Detector Claims · 23 October 2025 · secondary source
- Law360: 1st Circ. Skeptical That JPMorgan AI Tool Broke Mass. Law · 28 July 2026 · secondary source
2 · Safety
JPMorgan describes review and ongoing monitoring of models and a security programme, but red-teaming and independent audit reports are missing.
What is documented
- Model review challenges methodology and assumptions, can require additional testing and back-testing, and requires ongoing monitoring1
- An information security programme under Global Cybersecurity and Technology Control reports to the Audit Committee and the board1
- The CISO letter to suppliers names isolation of compromised suppliers and requirements for AI integrations3
- According to the Chief Data & Analytics Officer, data access and actions of AI agents are monitored7
The case against
Missing evidence of testing
- Test results, red-teaming and scope of validation for generative and agentic AI are not disclosed
- No independent audit reports on information security were found
Sources for Safety (9)
- J.P. Morgan Europe Limited (subsidiary, investor relations filing of JPMorgan Chase & Co.): J.P. Morgan Europe Limited, Annual Report for the year ended 31 December 2025 · 9 April 2026 · primary source
- American Banker: JPMorgan is working with U.S. regulators on generative AI pilot projects · 9 November 2023 · secondary source
- JPMorgan Chase & Co.: Patrick Opet, An open letter to third-party suppliers · 26 April 2025 · secondary source
- JPMorgan Chase & Co.: 2025 Sustainability Report · 3 November 2026 · secondary source
- Board Cybersecurity: JPMORGAN CHASE & CO 10-K Item 1C. Cybersecurity - 2026-02-13 · 13 February 2026 · secondary source
- JPMorgan Chase & Co.: Jamie Dimon, Letter to Shareholders (Annual Report 2025) · 6 April 2026 · secondary source
- American Banker: How banks are laying the foundation for agentic AI · 13 November 2025 · secondary source
- Board Cybersecurity: JPMorgan Chase & Co. Cybersecurity Incident (Tracker) · 6 February 2026 · secondary source
- AI Incident Database: Entity JPMorgan Chase · 10 April 2025 · secondary source
3 · Transparency
JPMorgan discloses its use of AI only in general terms and publishes neither system documentation nor explanations for customers and applicants.
What is documented
- The 2025 shareholder letter describes AI as shaping almost every function and acknowledges uncertainty and risks1
- J.P. Morgan AI Research and an Explainable AI Center of Excellence carry out research on explainability and fairness5 6
- The switch of proxy voting to the internal AI tool Proxy IQ was publicly announced3
The case against
Missing disclosure
No explanation to those affected
- Labelling and explanations of AI-supported decisions for customers and applicants are missing
Sources for Transparency (8)
- JPMorgan Chase & Co.: Jamie Dimon, Letter to Shareholders (Annual Report 2025) · 6 April 2026 · secondary source
- JPMorgan Chase & Co. (SEC EDGAR): Form 10-K for fiscal year 2025 (jpm-20251231.htm) · 13 February 2026 · primary source
- American Banker: JPMorgan drops proxy advisory firms in favor of AI platform · 7 January 2026 · secondary source
- Let's Data Science: JPMorgan Emphasizes AI, Risk Management, Record Growth · 13 April 2026 · secondary source
- J.P. Morgan: Explainable AI Center of Excellence · 3 November 2026 · secondary source
- J.P. Morgan: J.P. Morgan AI Research – About · 3 November 2026 · secondary source
- American Banker: JPMorgan Chase using advanced AI to detect fraud · 3 July 2023 · secondary source
- Evident Insights: Here's the 2025 Evident AI Index · 3 November 2026 · secondary source
4 · Why the harm rule does not apply
The harm rule does not apply: within the 36-month window there is no judgment, fine or pattern of serious incidents involving its own AI systems; the HireVue lawsuit is pending without a finding.
5 · Controversies (6)
- mediumAn applicant sued, claiming AI-evaluated HireVue interviews acted as a prohibited lie detector test; one claim was dismissed, the appeal is pending. · 28 April 2025 · open
- minorThe OECD AI Incidents Monitor records job cuts through AI automation; the CEO confirmed job losses and announced redeployment plans. · 24 February 2026 · unklar
- minorJPMorgan reported a data breach at a third party affecting an estimated fewer than 1,000 records. · 9 January 2026 · unklar
- minorThe fintech BondIT sued, claiming JPMorgan took its AI technology after partnership talks; JPMorgan has not commented. · 1 October 2025 · open
- minorThe CFPB sued JPMorgan and other banks over inadequate fraud protection on Zelle; the case was dropped in 2025. · 20 December 2024 · abgewiesen
- minorThe OCC fined the bank subsidiary USD 250 million for gaps in trade surveillance; the Federal Reserve imposed USD 98.2 million in parallel. · 14 March 2024 · abgeschlossen
6 · What remains open
- Wording of an AI policy or AI principles and their scope
- Risk classification and approval procedure for generative, agentic and purchased AI
- Results of validation, red-teaming and monitoring of the AI systems
- Group-wide description of cybersecurity governance in the 2025 10-K, which could not be read
- Outcome of the appeal Saint Cyr v. JPMorgan Chase & Co.
- Human oversight of the proxy voting tool Proxy IQ
7 · How it was assessed
Cut-off date: 3 November 2026 · 40 searches, 42 fetches.
Primary documents read (10)
- JPMorgan Chase & Co. Form 10-K 2025 (Part I to Item 1A, two filings)
- Proxy Statement 2026 and 2025 (first 100,000 characters each)
- Jamie Dimon, Letter to Shareholders 2025 (to p. 31)
- JPMorganChase 2025 Sustainability Report (extracted text)
- J.P. Morgan Europe Limited, Annual Report 2025 (incl. Estimations and Model Risk)
- OCC Consent Order AA-EC-2023-49
- CFPB notice on the Zelle lawsuit (with archival note on discontinuation)
- Brief of Appellant, Saint Cyr v. JPMorgan Chase & Co., First Circuit No. 25-2129
- CISO letter to suppliers (2025-04-26)
- Corporate pages on AI Research, Explainable AI Center of Excellence, AI and model risk governance, Leadership
Research of 3 November 2026 under Globalance AI Footprint rulebook version 0