Implementing AI RMF with Policy-as-Code Automation
summary
Maps Linux Foundation principles to NIST AI RMF and ships machine-readable controls plus working templates that enforce checks at CI/CD and Kubernetes admission points. Uses Open Policy Agent and Kyverno-style gates, outputs OSCAL artifacts via an OSCAL trestle pipeline from spreadsheet component definitions, and automates bias/privacy checks and generated LLM red‑teaming prompts. Implements a healthcare federated learning with VCG payoff that computes dollar penalties/rewards for bad or low‑quality contributions; deployed proof shows no cheating detected and continuous policy-as-code enforcement.
AI-written summary · about
chapters (12)
- 0:00AI risk management intro
- 3:06AI frameworks need shared language
- 7:23Threat modeling with AVGAI
- 10:57Red teaming LLMs
- 12:42Healthcare federated learning threats
- 18:22Game theory for AI policy
- 21:30NIST AI RMF manage automation
- 23:15Policy-as-code with OSCAL
- 27:50OPA policy validation
- 30:36OSCAL GitHub automation
- 33:01TAG AI risk compliance
- 34:04Q&A
related talks
30mRedefining Access Control: Scaling Policy as Code for Humans and AI Agents
30mCompliance at the Speed of Innovation: Leveraging AI-Driven Automation for Real-Time Regulatory Read
25mDe-risking Deployments in an AI World
12mUsing Privacy Infrastructure to Kickstart AI Governance: NIST AI Risk Management Case...
22mOSCAL in Action: Real World Examples of Automating Policy & Comp...
33mGoverning the Ungovernable: Security and Compliance for AI Agents in Open Source Projec...
12mThe AI Trust Triad for DevOps
11mKeynote: Automating RBI Compliance With Unified Policy-as-Code