Wayve pitches end-to-end 'Learned Safety' to build one AI driver for any vehicle
alexgkendall · x · 2026-11-06
Wayve CEO Alex Kendall argues that, as with driving itself, end-to-end deep learning offers a more generalisable and scalable representation for safety — the goal being an AI Driver that can safely drive any vehicle, anywhere. The stack includes the Wayve Learned Safety Envelope, closed-loop simulation with GAIA plus learned evaluation via GRADER, and grounding through language and reasoning. An accompanying graphic contrasts AV1.0's sensor-by-sensor metric code enforcing rules with AV2.0's single E2E neural network classifying behavior directly from raw input (safe? comfortable? road-rule compliant?).
More from Embodied
- RAX From Harvard/NVIDIA/Berkeley Lifts Robot Success From 26% to 94% With Frozen Weights — jiqizhixin · 2026-11-06
- Open-Weights Robot-Use Harness: Qwen3.8 Scores 31.39, Beating GPT-6 Astra's 28.97 — aigclink · 2026-11-06
- Dev builds Salvo battleship game for dual-screen iPhone Duo with head-to-head and AI play — HankYeomans · 2026-11-06
- First-hand demo of Wayve AI Driver on San Francisco streets impresses rider — Sethwinterroth · 2026-11-06
- YC backed 25 robotics startups last batch as VCs call robotics the new SaaS moment — Scobleizer · 2026-11-06
- Apple Vision Pro Positano scene hits 65fps on M2 with boats, yachts and 40 seagulls — Scobleizer · 2026-11-06