[ Intelligence ]
The Safest Miles. A Lasting Habit.
What 423 million miles reveal about the vehicles that almost never brake hard, and why the ranking still held six months later.
Read full article →
Most Counties Align. Twenty-Five Differ.
We measured 423 million miles across 1,223 US counties, each against its own driving mix. Only 25 were meaningfully different.
Read full article →
Why I Joined Empiric Earth
AI is leaving the screen. It will only go as far as people trust it, and in the physical world, trust is earned on safety, not demos.
Read full article →
The Study of Things That Did Not Happen
The debate over AI safety is asking the right question. For AI already operating in the physical world, we can start answering it.
Read full article →
Prime Day: Almost Nothing Moved
A shopping surge. No meaningful change in the measured fleet. We measured 842,800 rides across Prime Day and matched June weekdays.
Read full article →
The Margin for Error
Heat doesn’t make people drive worse. It shrinks the margin for error. We measured 1.5 million rides across three July heat waves.
Read full article →
Fewer Trips, Longer Journeys
Summer doesn’t put more cars on the road. It makes every trip longer. We compared 11.9 million summer rides against 12.3 million spring ones.
Read full article →
When AI Leaves the Screen
It will drive vehicles, route freight, inspect bridges, dispatch first responders, move robots through warehouses, and help cities operate. Its job will not be to produce information. Its job will be to understand the world well enough to act inside it.
Read full article →
What a Road Record Has to Prove
Empiric Earth launches today. Along with everything else a launch produces, here is the part that usually gets skipped: how to check whether any of it is true.
Read full article →
We Counted the Near Misses
When two vehicles collide, somebody is required to write it down. A police report, usually an insurance claim, sometimes a line in a federal database. Almost everything the world knows about road risk comes from those records.
Read full article →
Nexar and Nauto Are Now Empiric Earth
Read full article →
Nexar and Nauto to Merge, Creating the Independent Infrastructure Platform for Physical AI
Read full article →
The Physical World Does Not Have a Token Supply
Language models were built on a resource that turned out to be finite: the written internet. The physical world has no equivalent reservoir, and the reason why is the most under-discussed constraint in AI right now.
Read full article →
The Seconds Before the Crash
Every major advance in vehicle safety for a hundred years has been about what happens during impact. The seconds before it were treated as unreachable. That is no longer the case, and the industry has not caught up with what that means.
Read full article →
Nobody Should Grade Their Own Homework
Many companies building AI for the physical world evaluate their performance on their own data and using their own evaluation metrics. In any other field with this much at stake, that arrangement has a name and a regulator.
Read full article →
The DeltaWhat Simulation Cannot Teach You
A simulator will run a billion miles overnight for the cost of the compute. It is the best value in autonomy development and it is indispensable. It is also structurally incapable of the one thing the industry most needs from it.
Read full article →
