The Sequence Robotics - Issue #962: Rhoda AI and Robots That Predict Their Next Move
Was this email forwarded to you? Sign up here
The Sequence Robotics - Issue #962: Rhoda AI and Robots That Predict Their Next Move
How video prediction becomes a control policy and why touch changes the equation
TL;DR Rhoda AI is betting that predicting short sensory futures can organize robot behavior, with a separate inverse dynamics model turning each predicted future into movements. Its September scaling report ties better video prediction to stronger bearing-unpacking performance, and its October tactile report extends the loop to touch. The open question is whether these company-reported results, from one task and embodiment, carry over to other workflows, repeated training runs, and sustained factory operation.
Watch someone empty a cardboard box and your brain fills in what comes next. The lid opens, the bag rises, the contents fall. For a robot, however, recognizing that sequence leaves a rather inconvenient problem: which motors should move, and what happens when the bag snags? A convincing movie of successful work does not automatically become successful work.
Rhoda AI is building a company around that translation. Its bet is that predicting short sensory futures can organize robot behavior, while a separate model converts those futures into movements. The attraction is a vast supply of training videos. The challenge is making their lessons survive contact with an actual factory...
Subscribe to TheSequence to unlock the rest.
Become a paying subscriber of TheSequence to get access to this post and other subscriber-only content.
A subscription gets you:
© 2026 Jesus Rodriguez
75 Miracle Mile, Suite 7688, Coral Gables, FL 33134