Humanoid robotics enters commercial piloting as GR00T, Figure, and full-body sensing platforms converge on foundation-model control
NVIDIA's GR00T foundation model learns tasks from video and simulation; Figure 02 runs a built-in multimodal model for language, vision, and voice; the GENE.01 platform adds full-body skin sensing touch, proximity, force, and temperature. Goldman Sachs pegs the humanoid market at $38 billion by 2035, with 2026 the transition from prototype validation to commercial piloting.
The convergence is on architecture, not vendor. GR00T learning from video and simulation, Figure 02's built-in multimodal model, and Gemini Robotics 2's whole-body control are three routes to the same design: a transformer-based foundation model that maps multimodal input to joint control. When the leading platforms agree on the shape of the solution, the field has left the exploratory phase.
Full-body sensing is the underrated enabler. A humanoid that can perceive touch, proximity, force, and temperature across its whole body has the feedback needed for the contact-rich manipulation that real work requires — the difference between a robot that can pick up a rigid box and one that can handle something delicate or unexpected. GENE.01's skin is a bet that sensing, not just control, gates usefulness.
The $38 billion 2035 figure and the 'commercial piloting' framing set the honest expectation: 2026 is the year of paid pilots, not deployment at scale. The transition from prototype validation to piloting is real progress, but it is a beginning — the models and the bodies are ready to be tried in the field, and the field will decide which of them survive contact with actual work.
Meta Intelligence — Humanoid robots 2026: Tesla Optimus, Figure 02 & NVIDIA Isaac status → · Artic Sledge — AI humanoid robots 2026: technology, builders & future →