Wayve

Wayve's Brandon Basso demos how its self-driving AI learns what safe driving looks like, rather than relying on HD maps and external safety checkers: "The way that we've built the driving stack is with safety as an intrinsic property vs an extrinsic bolt-on property." "Lots of systems have high-definition maps that are really like checkers. What if the stop sign isn't there? What if it moved? What if there's a temporary stop sign? What if the stop sign is connected to a bus that's moving?" "The checker systems tend to be the ultimate bottleneck and limit of system performance." "What you see on the screen here is a learned representation of safety. You can think of this as like a future occupancy map. The red areas are the don’t drive in areas." "We use this system not as a bolt-on checker, but as an explicit head, a task that we learn, on how to drive safely." @wayve_ai

Wayve's Brandon Basso argues self-driving cars face the same fundamental AI safety challenge as OpenAI and Anthropic, except the consequences play out in the physical world: "The status quo today for driving is not great. It is a very risky thing that we do." "We have this novel approach to self-driving that we refer to as AV 2.0 or end-to-end systems, and there hasn’t been a lot of rigorous thinking yet about how we prove the safety of those systems." "It's exactly the same problem that folks at Anthropic and OpenAI think about when they're thinking about frontier safety. It’s the same problem but the physical interpretation of it." @wayve_ai