Lightning Talk: Your Prompt Is a Cross-Border Data Transfer
summary
Claims a per-request AI Receipt that traces prompts end-to-end using OpenTelemetry and the emerging GenAI semantic spec, showing when prompts touch vector DBs (RAG), MCP tool calls, or external models on GKE. Demo sends a local LLM request, surfaces region-level spans, tags each retrieval and external call, and flags unexpected cross-border hops; cites 2025 leakage rates and shows how on‑prem hosting plus span tagging converts sovereignty from a claim into verifiable evidence.
AI-written summary · about
related talks
01h 00mPrompt Failures and Latency Spikes: Observability for AI
29mYour Prompt Is Not a Unit Test — instrumenting LLM apps in production
09mLightning Talk: Bridging Silos: Real‑Time DevSecOps Telemetry for Container Def...
24mHow OpenTelemetry Helps Generative AI
09mGraduated Project Lightning Talk: Using Trace Data for Monitoring & Alerting of Appl...
21mLightning Talk: Distributed Tracing Without the Pain!
37mAI in SRE: Unlocking Prometheus Insights with Natural Language
01h 02mObserving AI Applications with OpenLit and OpenTelemetry