Brave New World: Welcoming New AI Identity Challenges
summary
Claims that AI creates a new programmable identity class and that verification must become multi‑dimensional ranking rather than binary allow/deny. Demonstrates an open‑source stack combining arcjet bot score 0–100, Opal policy engine, and lunard Dev egress proxy with RAG over a vector DB to filter authorized documents. Proposes proactive access requests, continuous access evaluation (CAPE) and feature‑flag driven session revocation; shows a working demo architecture integrating these pieces and argues it prevents unexpected agent actions while preserving useful automation.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:07AI trust in workflows
- 3:36Project context
- 4:21Reassessing identity security
- 7:34Authentication challenges
- 9:20Identity verification ranking
- 13:37Multi-dimensional identity ranking
- 16:04Authorization for AI agents
- 18:51Access control for AI systems
- 20:39Securing AI agent locations
- 22:29Retrieval Augmented Generation for security
- 26:05Authorization in RAG pipelines
- 27:08Session and token security
- 29:13Continuous access evaluation
- 31:23Summary
related talks
31mPanel: Trustworthy AI: Enhancing Human Agency and Machine Accountability in the Age of Agentic AI
04mYou Don't Need a New Identity Platform for the AI Era
20mCIAM for AI: Authn/Authz for Agents
40mIn AI We Trust? Securing the Future, One Agent at a Time
16mIdentity-Centric Security for AI Systems
25mAgents Are Not Users or Services: Securing AI Traffic with a New Identity Model
24mBuilding Trustworthy AI Authentication & Authorization for AI Agents
24mAI Governance for developers: A practical framework for deploying autonomous coding agents safely…