Agentic Security & Governance — From AI Safety to AI Readiness
Diagnose and govern autonomous AI agents at enterprise scale — anchored on the July 2026 Hugging Face incident. Overview + 4 hands-on challenges.
6 docs tagged with "agentic-security"
View all tagsChallenge 01 — Objective & Autonomy Risk
Reproduce the 'unintended path' safely, threat-model it with OWASP LLM06, constrain the objective, and map where human-approval brakes belong.
Challenge 02 — Permission & Blast Radius
Treat the agent like a digital employee: design least-privilege identity, cut standing permissions to Just-in-Time, and shrink the blast radius with a before/after diagram.
Challenge 03 — Data Protection & Runtime Monitoring
Protect data even after an agent gains access, and design runtime behavior monitoring that watches tool calls, access requests, and escalations — not just outputs.
Challenge 04 — Governance, Brakes & Executive Readout
Capstone: design human-approval gates, a kill-switch runbook, and access-revocation processes, then integrate 01–03 into a board-ready executive readout.
The Incident, in Depth — The July 2026 Hugging Face Agentic Attack
The full technical deep-dive on the July 2026 Hugging Face autonomous-AI incident: kill chain, risk generations, root-cause framework, maturity model, and primary-source evidence.