Waymo and DeepMind say they are collaborating to develop better self-driving AI algorithms, using techniques inspired by evolutionary biology
Kyle Wiggers / VentureBeat :
Context & Ripple Effects
This is the second documented bridge between Waymo and an Alphabet sibling lab: a year earlier, The Verge detailed how Waymo was already working with Google Brain researchers on deep learning and neural nets for its driverless vehicles. The DeepMind collaboration extends that pattern from applied neural nets into algorithm design itself, with evolutionary biology as the inspiration.
Seen against the rest of the coverage, the move reads as the early chapter of a long arc: by 2022 MIT Technology Review was profiling startups like Wayve, Waabi, and Autobrains betting on end-to-end AI learning to leapfrog Waymo, and by 2026 Waymo was pulling DeepMind's Genie 3 world models into training and building its own ASIC to cut Nvidia out. The 2019 announcement is where the lab-to-fleet pipeline became explicit.
First-order effects
- DeepMind's evolutionary-inspired techniques now flow directly into Waymo's driving stack, giving Waymo's researchers a second Alphabet lab alongside Google Brain to source methods from.
Second-order effects
- Rivals without an in-house frontier-research sibling — the end-to-end startups profiled by MIT Technology Review, and Cruise — face a competitor whose algorithm improvements are subsidized by Alphabet's broader research agenda rather than funded per-mile by robotaxi revenue.
Third-order effects
- If the pattern holds — Brain in 2018, DeepMind in 2019, Genie 3 and a proprietary ASIC by 2026 — Waymo consolidates into Alphabet's vertically integrated autonomy platform, where research labs, custom silicon, and the fleet itself form one closed loop that standalone AV startups must match piece by piece.
The trend: Alphabet is turning Waymo into the deployment vehicle for its research labs' output, with each new collaboration — from Google Brain to DeepMind to Genie 3 — tightening the loop between frontier AI research and the road.