Identify, assess, and document compliance risks specific to AI and agentic systems—including inaccurate or fabricated output, unauthorized or over-scoped agent action, inadequate consent and disclosure, unexplainable decisioning, disparate impact, data leakage across customers or organizations, prompt injection and third-party agent risk, and insufficient audit trail—and work cross-functionally to design pragmatic mitigation strategies. Working knowledge of the regulatory and supervisory framework relevant to AI in banking, including UDAAP, fair lending and ECOA/Reg B, FCRA and adverse action, Reg E and payments authorization, BSA/AML, GLBA and privacy, marketing and disclosure requirements, recordkeeping, complaint management, model risk management expectations, and third-party and vendor risk management guidance.