How does CIBC Bank USA govern responsible AI?
AI governance at CIBC Bank USA is set by its Toronto parent. The board's Technology Committee reviews CIBC's Data and Artificial Intelligence strategies at least annually, per its mandate, and the 2025 annual report says the new committee oversees the technology, data and AI strategy. Section 5.4 of the Code of Conduct effective November 2026, filed with the SEC on November 2, 2026, says staff use AI 'safely and responsibly', guided by CIBC's Trustworthy AI Principles. In March 2025 CIBC became the first major Canadian bank to sign the federal generative-AI code of conduct, citing its Enterprise AI Framework and AI Risk Assessment Process. AI and machine-learning models sit in the model inventory ahead of OSFI Guideline E-23 (May 2027), and the Q3 2026 risk section names frontier models such as Claude Mythos as a cyber risk. The 2025 Sustainability Report says the boards of CIBC Bancorp USA Inc. and CIBC Bank USA received education on AI and automation. For agents, CIBC's July 2026 position is 'agentic AI with humans in control', and CAI 2.0 began with about 250 handpicked testers. No US-specific AI policy or AI officer for governance is published.[1][2][3][4][5][6][7][8]
Data as of Nov 17, 2026 · CIBC profile updated Oct 26, 2026 · 14 sources cited on this page · 141 banks profiled on this site · Markdown
Which policies and bodies govern AI at CIBC?
| Item | What it says | Scope | Date |
|---|---|---|---|
| Technology Committee mandate | Reviews 'CIBC's Data and Artificial Intelligence ("AI") strategies' and approves AI investments; reviews appointments of EVP-level technology, data, AI or cybersecurity executives | Board, group and subsidiaries | 2026-06-10[1] |
| Code of Conduct, section 5.4 | 'Guided by CIBC's Trustworthy AI Principles, we are committed to the safe and responsible use of AI'; staff must follow policies when developing, procuring, deploying and utilizing AI | All team members | 2026-11-02[3] |
| Government of Canada voluntary generative-AI code | First major Canadian bank to sign; cites Trustworthy AI Principles, Enterprise AI Framework and AI Risk Assessment Process | Group | 2025-03-25[4] |
| 2025 annual report | Board Technology Committee oversees the technology, data and AI strategy; AI Governance and Risk Assessment Committees; AI and ML models in the model inventory | Group | 2025-12-04[2] |
| Q3 2026 Report to Shareholders | Names 'Claude Mythos and frontier models' and AI-driven phishing as cyber risks; model-risk work ahead of OSFI E-23 | Group | 2026-08-27[5] |
| 2025 Sustainability Report | Boards of CIBC Bancorp USA Inc. and CIBC Bank USA received education on AI and automation; Enterprise AI Governance named among teams | US subsidiaries | 2026-03-20[6] |
How does CIBC apply those controls to its AI tools?
CAI access requires a mandatory training course; the platform launched bank-wide on May 27, 2025.[9]
CIBC chose to build its own Teams meeting note-taker because giving a third party access to meeting audio was 'a non-starter'.[10]
CAI 2.0 started with about 250 handpicked employees; Jaime Tatis said 'It's too powerful to have 50,000 people just using it without control.'[8][11]
The July 2026 essay 'Agentic AI With Humans In Control' puts human judgment, governance and culture at the centre of agent design.[7]
In 2025 CIBC engaged with the Government of Canada on responsible AI development and deployment, and the board received director education on agentic AI.[6]
What does the CIBC profile record on this topic?
The rows below are the entries of the CIBC AI strategy profile that bear on this page, as of Oct 26, 2026; the profile holds the full record of 30 dated moves, 17 use cases and 40 sources.
| Use case | What the record shows | Status | Source |
|---|---|---|---|
| AI and ML in the model inventory; AI governance committees | Model-risk governance covers AI and ML models ahead of OSFI Guideline E-23 (May 2027); enterprise AI Governance and Risk Assessment Committees; third-party risk management covers AI-related risks. | In production | [2][5] |
| Frontier-model cyber risk | The Q3 2026 risk section names advanced models 'including Claude Mythos and frontier models' and AI-driven phishing, vulnerability discovery and autonomous attack chains, with faster patching in response. | In production | [5] |
| 'Agentic AI with humans in control' | Human judgment, governance and culture at the centre of agent design. | In production | [7] |
| Azure-hosted data and a ChatGPT-based CAI | Most consumable data sits in Microsoft Azure with Purview for governance; Forbes described CAI as based on ChatGPT. | In production | [12][13] |
| Date | Move | Detail | Source |
|---|---|---|---|
| Aug 20, 2026 | AI-experienced director appointed to the CIBC board | Prasanna Gopalakrishnan, formerly ADP's Global Chief Product and AI Officer, joins from 1 September 2026. | [14] |
| Jul 20, 2026 | Inside the CAI platform; agentic AI 'with humans in control' | A home-built meeting note-taker, fluency bell curves, and a governance-first agent stance. | [10][7] |
| Dec 4, 2025 | 2025 annual report: CAI across the bank, board oversight of AI, US data and technology work | CAI launched bank-wide with required training for every team member; a new board Technology Committee oversees the technology, data and AI strategy; the US segment lists a new wires platform, CRM tools, enhanced AML capabilities and better risk analytics. | [2] |
| Mar 25, 2025 | First major Canadian bank to sign the federal generative-AI code of conduct | CIBC cites its Trustworthy AI Principles, Enterprise AI Framework and AI Risk Assessment Process; 2024 pilots were CAI, Knowledge Central and GitHub Copilot. | [4] |
Does CIBC have AI principles?
Yes. CIBC's Trustworthy AI Principles are cited in section 5.4 of its Code of Conduct (November 2026) and in its March 2025 signing of the Government of Canada's voluntary generative-AI code. The full principles text is not published as a standalone document.
Who oversees AI at CIBC's board?
The Technology Committee. Its mandate has it review CIBC's Data and AI strategies at least annually, approve AI investments, and review appointments of executive vice-presidents in technology, data, AI or cybersecurity.
Is AI governed separately at CIBC Bank USA?
Not that CIBC has published. Policies are group-wide; the 2025 Sustainability Report says the boards of CIBC Bancorp USA Inc. and CIBC Bank USA received education on AI and automation.
Which AI risks does CIBC disclose?
Its Q3 2026 report names frontier models such as Claude Mythos and AI-driven phishing, vulnerability discovery and attack chains as cyber risks, and covers AI and ML models under model-risk governance ahead of OSFI Guideline E-23.
- Canadian Imperial Bank of Commerce: Technology Committee Mandate
- CIBC Annual Report 2025
- CIBC Code of Conduct, November 2026 (Form 6-K, Exhibit 99.1)
- CIBC becomes the first major Canadian bank to sign the federal generative AI code of conduct
- CIBC Report to Shareholders for the Third Quarter, 2026
- CIBC 2025 Sustainability Report
- Agentic AI With Humans In Control
- CIBC Curbs How Many Staff Can Test Out New AI Tool in Cost Focus
- CIBC launches in-house AI platform enterprise-wide after successful pilot
- Scaling AI at CIBC: Inside the CAI Platform
- CIBC weighs costs as select employees test more powerful AI tool
- Modern banking, reinvented: CIBC's blueprint for successful AI transformation
- CIBC CIO Richard Jardim On Building A Bank For The Modern World
- CIBC Appoints Prasanna Gopalakrishnan to its Board of Directors
When CIBC 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