The AI PC and the Era of the Modern Web
summary
Intel pushes AI onto local devices with new silicon and browser hooks and claims near‑native ML speed using Web Neural Network plus WebGPU and WebAssembly SIMD. Evidence includes VSCode and Photoshop running in browser, "65% of CPU cycles" on web workloads, and conferencing optimizations that yield 20% longer battery life. Demoed MPU usage, compute‑pressure states (nominal→critical), and a timing side‑channel exploit that required Chrome mitigations, so APIs work but privacy and model download size remain practical constraints.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:50AI moving to local devices
- 3:58AI PC and hardware
- 6:21Why the web matters for AI and modern computing
- 8:07AI web app building blocks
- 11:38SIMD and WebAssembly optimizations
- 14:05AI on the web platform
- 17:40Web GPU for AI
- 21:32AI APIs for video effects
- 26:25Web Neural Network API
- 29:21Compute pressure API
- 31:52Privacy challenges
- 34:02Summary and future
- 37:16Q&A
related talks
30mBeyond the Web of Today
18mWeb AI on next generation AI PCs
41mEscaping the Cloud: High-Performance AI in your Browser
24mWeb Apps of the Future With Web AI
46mThe Ultimate Guide to Browser-Based AI: Privacy-First, Offline-Ready and Future-Proof
12mBrowsing, building, and web in the AI era
16mAccelerating Web AI on
16mThe Web is AI Ready—maximize your AI web development with WebNN