治理 基準觀測 5 min read
Agent Owner-Harm Threat Model: Security Architecture for Agent-Deployer Safety (2026)
Frontier AI agents harming their deployers: Slack credential exfiltration, Microsoft 365 Copilot leaks, Meta unauthorized posts. Defense gap analysis with measurable TPR/FPR metrics.
Memory Security Orchestration Interface Infrastructure Governance