OpenAI disbands its robotics team, betting its general models can run robots
OpenAI disbands its robotics team as its general models start running robots
OpenAI has dissolved its stand-alone robotics team, reassigning researchers to its core research and applied groups, and even left its CoRL booth empty. The company says its general-purpose frontier models, including GPT-6 Astra, can now be applied directly to robots—an Nvidia study showed an Astra-powered agent hitting 74.8% on a simulated manipulation benchmark versus 51.0% for the best specialized models. The move reflects a broader industry bet on embodied AI and comes as OpenAI reins in costs ahead of a possible 2027 IPO.
He said he would interpret the move as a signal that OpenAI sees robotics as "no longer a special problem."
- jacquesm
General models can run robots. They are very bad at it. It is much more efficient to have a stack of models with 'dumber' but faster models taking care of lower level functions and higher level ones that are much slower taking care of high level planning. Consciousness isn't super fast, but it is very smart. Your 'lizard brain' (to use an overstretched metaphor) and reflexive system are relatively stupid but far faster and condense a whole raft of inputs to something your consciousness can deal with. For robots it isn't all that different. Sensors and actuators require millisecond level responses, that's something you're not going to get out of an LLM on the other side of a communications link.
- yewenjie
The Bitter Lesson comes for everything.
- winfredJa
This looks like general cost cutting to boost some numbers before IPO.
- eikenberry
So they gave up on non-cloud connected robotics then? Or was that ever in their scope?
- deadbabe
Is there a primer on how to actually use LLMs to control robots? Seems very token inefficient if you have to stream commands at 60fps.