Waymo co-CEO Dmitri Dolgov laid out the clearest technical case yet for why cameras alone can’t take a self-driving system to full autonomy, arguing that “weak sensing” hits a safety ceiling long before it reaches superhuman performance.
He never said the word Tesla. But camera-only is Tesla’s entire bet, and this was a direct shot at it.
The sensor debate, from someone with 220 million driverless miles
Dolgov made the comments in a talk at Y Combinator’s Startup School, walking through the lessons Waymo has learned building its driver over close to two decades. He put the sensor question on the table plainly: “there’s been a long-standing debate about what kind of sensors do you actually need for autonomous driving.”
His answer draws the line that camera-only advocates tend to skip right past. “Humans of course can drive with just eyes, so there’s that proof of existence,” he said. “If the goal were to just approximately match human performance or to build an assist product, that’s a very reasonable way to go.”
Then the catch. If you’re targeting full autonomy and strongly superhuman performance, he said, “you find that weak sensing just leads to a safety curve that flattens out way too early.”
That’s the whole argument in a couple of sentences. Cameras can match a human. They can power a driver-assist product. Dolgov’s point is that they top out below the bar a car has to clear to drive with nobody behind the wheel.
Unless you do it for demonstration purposes in a geo-fenced area with limited speed and low availability to limit danger through less mileage.
Why Waymo runs cameras, lidar, and radar together
Waymo uses three sensing types, and Dolgov spent real time on why. Cameras give you high resolution and color, but they’re passive, and they degrade in darkness and glare. Lidar directly measures the 3D structure of the world. Radar punches through fog, rain, and snow, and reads velocity directly with Doppler. Lidar and radar are active sensors, so they see just as well in pitch darkness or straight into a blinding sunset.
This isn’t redundancy for its own sake. “These different sensing modalities, they’re not backups to each other,” Dolgov said. Each one runs its own encoder, and the data fuses into a single view of the world that he says is “vastly superior to what you get with any one sensor.”
He backed it with cases where a camera goes blind. A dust storm in Phoenix, where the camera sees almost nothing while lidar cleanly picks out a pedestrian at the roadside. People hopping a concrete barrier onto a road at night. Kids chasing dogs across a pitch-black street with no headlights or streetlights on them. In each one, the camera feed is close to useless and the lidar view is clean.
Then there’s the failure case camera-only can’t engineer its way around: something physically covering the lens. A single leaf on a sensor can bring a robot to a full stop, Dolgov said, before showing a Waymo that caught a tree branch its wipers couldn’t shake and used its other sensors to drive itself safely back to the depot.
The ‘nines’ problem
Under the sensor argument sits a math argument, and it’s the part that should worry anyone extrapolating from a camera-only demo. Reliability lives on what Dolgov called an “exponential ladder of nines.” Tesla CEO Elon Musk calls it the “march of the nines.” Getting to 90% or 99% is the easy part. Every additional nine of reliability takes roughly ten times more effort than the one before it.
So the trap is picking the technology with the fastest early ramp, the one that makes the best demo, projecting that steep slope forward, and then hitting a plateau “way before the performance that is required by your product.” A demo might need one nine. A driver-assist product needs a few. A car with kids in the back and nobody driving needs a whole stack of them.
Dolgov also pushed back on the one knock that camera-only fans always reach for: cost. Waymo is on its sixth generation of hardware, and he said each generation has drastically cut the price. Betting against sensors like lidar on today’s prices, he warned, means betting on “a number that has a fairly short shelf life.”
Where Tesla fits
Tesla is the loudest bet on the other side of this debate. It pulled radar from its cars in 2021 and dropped ultrasonic sensors in 2022, going all-in on cameras with “Tesla Vision.” Elon Musk has called lidar “a fool’s errand” and said anyone relying on it is “doomed.” Tesla’s robotaxi service, which launched in Austin in 2025, runs on cameras alone. We’ve covered the Tesla Vision versus lidar fight in depth before.
So far the results line up with the exact ceiling Dolgov described. Tesla’s own robotaxi data shows a crash rate about three times worse than human drivers, even with a human safety monitor in the front seat. Musk has claimed FSD is safer than humans, release after release, and the data hasn’t backed him up yet, even after being heavily massaged by Tesla.
Waymo is running the numbers Dolgov says camera-only can’t reach. Its latest safety report, based on 220 million rider-only miles, shows 94% fewer serious-injury crashes than human drivers would cause over the same distance, roughly 17 times better. Waymo is now serving around half a million paid trips a week across 15 US cities and is scaling toward a 1-million-weekly-rides target, even as Musk insists Waymo “never had a chance” against Tesla.
Tesla recently confirmed it has accumulated 380,000 driverless miles over the past year since launching its robotaxi. Waymo’s driverless service does that in a day.
Author: Fred Lambert
Source: Electrek
Reviewed By: Editorial Team
