MOTION INTELLIGENCE MODEL
The missing element in physical AI.
Language models reason about information. MIM reasons about motion. No GPS. No cloud. No pre‑built maps.
01
What MIM is.
A foundation model for physical motion. It lets any machine know where it is, where everything around it is, and what is about to happen. In real time, on the machine itself.
Measured for the U.S. Army, the U.S. DOT and the top three automakers. Paid for by BHP, PACCAR, the U.S. Army and DARPA.
More done per operator.
50% FASTER MISSIONS · 67% LESS WORKLOADU.S. ARMY TRADOC See the numbers SafetyFewer crashes.
0 VS 4–6 COLLISIONS, GOVERNMENT‑RUNU.S. ARMY TRADOC See the numbers SecurityWorks when GPS doesn't.
±10 CM WITH GPS BLOCKED, BY THE VEHICLE’S OWN ESTIMATEU.S. ARMY See the numbers02
How it works.
Position comes from the network itself.
Machines measure their distance to each other, up to 1,800 meters apart, even with no line of sight between them. Turn GPS off and the machines still know where they are to within about ten centimeters.
See it work
~10 CM · NO LINE OF SIGHT MACHINE TO MACHINE
✕ GPS
03
Platform.
One model. Six capabilities.
From a truck yard to a city corridor.
04
Industries.
Where it already works.
Mining A working copper mine, and a second operator on its own account.
TWO PAYING BUYERSBHP · MIOS See the deployment
Logistics A heavy-truck yard, and underground rail tracked from anchors (small fixed radio beacons) in the stations alone.
~10 CM IN A HEAVY-TRUCK YARDPACCAR See the deployment
Defense Mine-clearing lanes, through to a long-term program funded by Congress.
130 / 135 MINES MARKEDU.S. ARMY See the deployment05
The team.
Robotics, DARPA, and three exits.
David Bruemmer Founder & CEO 
Phil Mann COO 
Michael Nelson President, Federal 
Taryn Rose, MD Co-founder & CSO