Building Sovereignty: Where Your AI Runs Matters
summary
Defines digital sovereignty as retaining effective control over critical data and workloads and demonstrates a workload‑first assessment that converts policy objectives into engineering artifacts using a dependency inventory graph and component placements. Uses a concrete seven-day hospital triage mission for clinical AI inference, enumerates protected assets (GPUs, model weights, patient data), runs a scenario test and finds an IM credential cache 24h expiry that breaks the continue power, and recommends strengthening exit capability or local hosting and rehosting procedures.
AI-written summary · about
chapters (6)
- 0:00Digital sovereignty in AI
- 3:54Sovereignty and system architecture
- 7:33Sovereign workload archetypes
- 12:35Sovereignty as workload control
- 17:34Workload sovereignty analysis
- 22:36Sovereignty and engineering takeaways
related talks
33mAI sovereignty, provenance and compliance
20mWrap-Up Session Panel Discussion: Sovereign AI
27mSovereign AI: Building...
42mInnovate Now, Secure Later? Decisions, Decisions…
09mSecuring Generative AI Workloads
24mSovereign Data for AI with Python
36mAvoiding 5 missteps that undermine your AI readiness and success
33mAI in the Security Stack