Lead security risk assessments for AI/ML pipelines, generative AI deployments, and agentic AI systems, addressing risks such as prompt injection, data leakage, model poisoning, excessive agency, insecure tool/plugin integration, and supply chain risk in third-party models. Partner with data protection, identity, and vulnerability management teams to extend existing enterprise security programs into AI-specific contexts — including access controls for model endpoints, data classification for training/inference data, and secure API/agent orchestration.