UMA Unveils First Humanoid Robot with Real-Time Learning AI

UMA has unveiled the design of its first humanoid robot alongside Real-Time Learning, a new AI architecture that enables robots to acquire new skills through human demonstrations rather than manual programming. Announced during the Machina Summit in Paris, the new platform is designed to accelerate deployment of intelligent humanoid robots across factories, warehouses, logistics centers and industrial facilities.

Founded in 2025 by former leaders from Tesla, Google DeepMind, NVIDIA and Hugging Face, UMA is developing Physical AI systems that help industries address growing labor shortages. The company believes humanoid robots can integrate naturally into workplaces built for people while performing repetitive, physically demanding and hazardous tasks safely alongside human workers.

Rémi Cadène, CEO and Co-founder of UMA, stated:

"Demographic, industrial and environmental challenges all point to the same reality: societies need greater productive capacity. We believe intelligent robots will become part of the solution, not as a substitute for people, but as a new class of tools that enables them to devote more time to creativity, judgment, innovation and caring for others."

At the core of UMA's platform is Real-Time Learning, which allows robots to observe demonstrations, practice tasks and continuously improve through experience instead of requiring engineers to program every workflow. The company says this approach significantly increases deployment flexibility while reducing engineering complexity across diverse industrial environments.

Rémi Cadène added:

"The industry's challenge is no longer building robots capable of executing individual tasks. The next breakthrough lies in building robots capable of learning new ones."

Headquartered in Paris, with teams in London and Geneva, UMA plans to continue advancing its Physical AI platform for large-scale industrial deployment. The company believes Europe is well positioned to lead the next generation of humanoid robotics by combining world-class research, industrial expertise and increasing demand for intelligent automation.