# Treasury Nominee Leaves AI-Agent Liability Unresolved

Treasury nominee Rebekah Jurata declined to define who would be liable or what guardrails should apply when unauthorized AI agents act in bank accounts, citing unresolved technical and policy questions. The issue signals emerging federal attention to AI security and fraud controls in financial services, but the signal identifies no solicitation, contract award, or specific procurement requirement.

- Firms providing AI assurance, cybersecurity, or financial-sector risk management can use this as an early policy signal when shaping relevant offerings; any future requirements remain undefined.
- Contractors serving Treasury or financial regulators should account for unresolved questions around agent authorization, accountability, and fraud controls in financial-sector work.
- Because no procurement opportunity or compliance deadline is identified, businesses should treat this as policy context rather than a near-term bid opportunity and look to subsequent Treasury guidance or regulatory activity for specifics.

**Jurisdictions:** federal
**Industries:** Information Technology
**Topics:** Artificial Intelligence, Cybersecurity
**Published:** November 20, 2026

### Government Entities
- U.S. Department of the Treasury (Treasury)
- Senate Committee on Banking, Housing, and Urban Affairs (Senate Banking Committee)
- Securities and Exchange Commission (SEC)

### Sources
- [Treasury nominee noncommittal on responsibility for rogue AI agents in accounts | FedScoop](https://fedscoop.com/treasury-nominee-ai-agents-bank-accounts) - FedScoop