Privacy-Preservation with Smarter Automotive Experiences
summary
Claims a working platform that keeps raw data in-car and ships only model weights or zero knowledge proofs, deploying Wasm modules to both Nvidia and Intel stacks and running training on-device across an orchestrated fleet of 18,400 vehicles and edge POPs in over 100 data centers. Components named: Tyra (privacy wallet), Ray (responsible-AI toolkit), Namio (graph data space). Runtime choices: wasmTime, Wasm micro runtime; delivery via Fastly edge network. Results: moved some OEMs from costly data collection toward zero-data analytics.
AI-written summary · about
chapters (8)
- 0:00Privacy in automotive
- 3:10Automotive data privacy
- 7:44Tyra Ray and Namio tools
- 12:39Vehicle architecture for privacy
- 16:07WebAssembly for embedded
- 22:18Ray toolkit for fleet training
- 27:52Privacy-first inference benefits
- 30:40Q&A on OTA updates
related talks
39mFrom Event-Driven to Automotive
30mProtecting Wasm Workloads with Confidential Computing and TEE...
08mLightning Talk: Build trusted cloud apps with Wasm: WebAssembly Micro Runtime is ready
25mLeveraging Wasm for Portable AI Inference Across GPUs, CPUs, OS & Cloud-Nativ...
40mWebAssembly for AI Infra : A Lightweight, Fast, and Secure Alternative Approach
38mStreamlining AI Solutions Development: S...
19mWebAssembly's Journey Beyond the Browser
13mKeynote: WEdge Project: Edge AI enabled by Wasm