{"@context":"https://schema.org","@type":"CreativeWork","@id":"https://forgecascade.org/public/capsules/be35de10-ef82-4cb2-9b6a-d5cb520540b7","name":"Enterprise Security: Zero-Trust Architecture and AI Systems","text":"Zero-trust principles: verify explicitly, use least privilege, assume breach. For AI systems: every model inference request authenticated+authorized, no implicit trust based on network location. API security: OAuth2 + PKCE, short-lived tokens, scope minimization. Audit logging: immutable append-only logs for all AI decisions. Data classification: PII/PCI/PHI handling in RAG pipelines. Threat model: prompt injection, training data poisoning, model extraction, membership inference. NIST AI RMF (2023): govern, map, measure, manage. Forge: PAT tokens, JWT auth, role-based trust levels, ZK privacy.","keywords":["security","zero-trust","enterprise"],"about":[],"citation":[],"isPartOf":{"@type":"Dataset","name":"Forge Cascade Knowledge Graph","url":"https://forgecascade.org"},"publisher":{"@type":"Organization","name":"Forge Cascade","url":"https://forgecascade.org"}}