Identity + AI = Intelligent Applications
summary
Demo shows an end to end system tying identity, encrypted user data and on device AI to mental healthcare workflows using Ory for auth, Supabase for vectors and a prototype IPFS node that streams MRIs and shaves a 35 minute manual process. User context lives in encrypted SQLite tied to passkeys and decentralized IDs, fed into a LangChain RAG pipeline with vector search and local fine tuning that runs in hours and redistributes models via IPFS using Ory + Supabase + IPFS and edge quantized Zephyr models for offline resilience and personalization.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker welcome
- 1:10About the speaker and Tribe Health Solutions
- 3:00Core platform with Auth0 and Supabase
- 4:45Decentralized data with IPFS
- 6:36Mental health use case and personal story
- 9:30Generative AI with user context
- 13:03Future of smaller AI models
- 16:53Self-hosting AI models and resilience
- 18:19Decentralized healthcare and data sharing
- 21:35Q&A and closing remarks
related talks
33mIAM, Agent: Identity for Autonomous AI
26mAI-first frontend architecture
23mHow to overcome AI threat to user authenticity
24mAI, Health & Confidentiality
16mIdentity-Centric Security for AI Systems
35mIdentity and Access Management for the Decade Ahead
04mYou Don't Need a New Identity Platform for the AI Era
37mPanel Discussion: Start Simple to Scale Decentralized Identity