Test the security of AI-enabled applications and agentic systems (AI red teaming), including prompt injection, jailbreaks, insecure output handling, excessive agency, tool and plugin abuse, data leakage, and model and supply chain risks, aligned with the OWASP Top 10 for LLM Applications and MITRE ATLAS; and apply AI-assisted techniques to enhance testing. GCRS delivers across four service lines: Cyber Risk Advisory: program maturity assessments (ISO 27001, NIST, CIS Controls), risk-based program prioritization, security and privacy governance, data privacy and cyber compliance (GDPR, CCPA, NYDFS, HIPAA, PCI-DSS), third-party risk, AI security assessments, and board reporting.