FINRA 2026 Annual Regulatory Oversight Report: generative AI and AI agents
Names the risks of AI agents at member firms (acting without human validation, acting beyond the user's intended scope and authority, hard-to-trace multi-step reasoning, mishandling sensitive data, misaligned reward functions) and encourages agent-specific supervision: monitoring agents' system and data access, deciding where a human must review, tracking agent actions and decisions, and guardrails that constrain behavior, within existing duties on supervision, communications, recordkeeping, and fair dealing.
Draft mapping. Drafted by the Register's agents from the rule's text and not yet confirmed by a person. It is the Register's reading, not legal advice: whether and how the rule applies to you is a question for counsel.
Controls this rule maps to
Required by the text The rule's text requires a measure that is this control, or plainly includes it. Implied A binding rule whose purpose needs this control, but whose text does not name it. The Register's reading. Recommended A voluntary standard or guidance recommends it; nothing binds you to it. Proposed Appears in a proposed rule or proposed provision that is not yet in force.
| Control | Strength | Where in the rule, and why | Confidence | Missing or failed in |
|---|---|---|---|---|
| Machine-enforceable scope C-M2 | Recommended | guardrails that constrain agent behavior to the user's actual and intended scope and authority | medium | 4 cases |
| Least-privilege credential scoped to task C-E1 | Recommended | monitoring and limiting what systems and data an agent can reach | medium | 5 cases |
| Human takeover with a time window C-R3 | Recommended | deciding where human-in-the-loop review is required before an agent's action stands | medium | 2 cases |
| Independent audit log C-V1 | Recommended | tracking agent actions and decisions so outcomes can be traced | medium | 1 case |