AI On The Farthest Edge
summary
Claims include 10,000× power efficiency from neuromorphic chips, running containerized models on microcontrollers via lightweight containers, and microjoules per inference with millisecond latency. Demonstrations span a peel‑and‑stick intelligent smoke detector that counts occupants for 5‑year battery life, vibration sensors moving from yearly to five‑year operation, and payload/flight optimizations that extend drones 60–80%. Panelists describe what was built, the orchestration tools used, and concrete deployment numbers that actually work.
AI-written summary · about
chapters (6)
- 0:00Welcome to farthest edge AI
- 1:04Panelist introductions
- 3:33Defining farthest edge AI
- 9:57Deployments with farthest edge AI
- 17:38AI hype for startups
- 21:13Closing remarks
related talks
28mAI vs Climate Change: Insights from the Cutting Edge
31mKeynote: The next big thing in machine learning is tiny
22mAdvancing the Human Experience
22mExtending European Strength with Embedded AI
48mThe Power of Edge AI: Revolutionizing Future of Real-Time, Autonomous Processing
21mA New Architecture for the Intelligent Edge
36mApplied AI on the edge
29mEdge Intelligence Revolution