TickerTain
$SPCX····$NVDA····$MSFT····$GOOGL····$ANTHROPIC·$TSLA····$CRWV····$NBIS····$HIMS····$TSM····$ASML····$FIGURE·$LYFT····$UBER····$SNOW····$META····$ELEVENLABS·$CURSOR·$COGNITION·$OPENAI·$AAPL····$RKLB····$XAI·$AMZN····$AMD····$000660.KS····$INTC····
← newsroom
Robotics & Physical AI · Physical AI requires solving four fundamental gaps: cost of error, latency, data, validation
now playing · Robotics & Physical AI
AI Infrastructuretailwindscore 10/10dmitri dolgov
Three-AI flywheel (agent, simulator, critic) powered by shared foundation model; eval/metrics are the strategic moat
Dolgov describes a flywheel where real-world deployment generates data → grounds simulator → simulator generates hard cases for critic → critic scores and improves agent → smarter agent dep…
Autonomous Vehiclestailwindscore 9/10dmitri dolgov
Waymo proves AVs have crossed the chasm: 17x safety, exponential scaling, 15-city deployment
Autonomous vehicles have moved from demo to product at scale — Waymo's 220M+ miles, 17x safety advantage, and 4M weekly miles demonstrate the technology is commercially viable and entering…
Robotics & Physical AItailwindscore 9/10dmitri dolgov
Physical AI requires solving four fundamental gaps: cost of error, latency, data, validation
Deploying AI in the physical world faces four structural gaps vs digital AI: irreversible cost of errors (human lives), millisecond latency constraints, no internet-scale pre-labeled data,…
Autonomous Vehiclestailwindscore 9/10dmitri dolgov
Waymo CEO: Demo is 1% of work; reliability requires exponential effort per nine
Dolgov argues that achieving full autonomy requires climbing an exponential ladder of reliability 'nines' where each additional nine takes 10x more effort, and that every AI breakthrough ma…
Autonomous Vehiclestailwindscore 9/10dmitri dolgov
Waymo bets on multi-modal sensing (camera+lidar+radar) for superhuman safety; camera-only flattens too early
Dolgov argues that camera-only sensing hits a performance ceiling far below superhuman levels required for full autonomy, and that fusing cameras, lidar, and radar provides complementary ph…
AI Infrastructuretailwindscore 9/10dmitri dolgov
Waymo Foundation Model: multimodal world-action-language model with System 1/2 architecture for physical AI
Dolgov describes Waymo's foundation model as a multimodal (camera/lidar/radar), world model (physics + social semantics), action model (understands agent's effects), language-aligned (unloc…
AI Infrastructuretailwindscore 9/10dmitri dolgov
Closed-loop simulation with generative world models enables training on synthetic rare events never seen in real world
Dolgov argues that closed-loop simulation (where agent acts, sees world response, acts again) is absolutely vital for safety-critical physical AI, and that building a high-fidelity generati…
AI Infrastructuretailwindscore 9/10dmitri dolgov
Eval and metrics are the strategic moat in physical AI — not model architecture
In safety-critical physical AI, the defensible advantage lies in evidence-grade evaluation frameworks and closed-loop simulation flywheels (agent, simulator, critic) that compound real-worl…
Autonomous Vehiclestailwindscore 9/10dmitri dolgov
Waymo demonstrates 17x safer than human drivers on serious injury crashes over 220M autonomous miles
Dolgov presents Waymo's latest safety data: 17x better than human drivers on serious-injury crashes across 220M+ fully autonomous miles, preventing a serious injury every 8 days — arguing t…
AI Infrastructuretailwindscore 8/10dmitri dolgov
Structure-augmented end-to-end models channel scale; vanilla end-to-end fights scale in physical AI
Dolgov applies Sutton's 'bitter lesson' to argue that structure which fights scale loses, but structure that channels scale (like physics, rules of road, object behaviors) wins — Waymo's 's…