Farming Robots' Toughest Problem Is Mud, Not AI — Engineers Build World Models to Sense It

Scobleizer · x · 2026-08-15

Robert Scoble visited Reservoir Farms, where an ag-robotics company told him the biggest challenge in outdoor environments isn't perception or navigation—it's simply not getting your robot stuck in the mud. His takeaway: these teams must be building some really good world models for sensing mud. "Imagine telling your parents that's your job," he joked—yet another dirty job for the AI era.

The quoted thread announces Ruggedize, a first-of-its-kind deep tech conference for ag robotics and the physical AI stack: August 26–27, 2026 in Salinas, CA, with speakers including Google DeepMind DevRel lead Paige Bailey, plus NVIDIA, John Deere—and, really, the FBI—sharing how their tech is used in the fields.

Six core focus areas from the site:

Original post →

More from Companies & People

Companies & People channel →