UMA's founders were instrumental in shaping the last decade of breakthroughs in deep learning, robotics and open-source AI. They are convinced that the next era of artificial intelligence won't unfold on screens, it will happen in warehouses, hospitals, labs, factories and homes, where machines must face friction, unpredictability and human complexity. UMA is built for this transition: from digital intelligence to physical autonomy.

AI Is Moving Off the Screen and into the World

The past ten years gave rise to generative models, multimodal systems and language-based intelligence. The next decade will be defined by robotics, powered by an AI that can see, move, manipulate and make decisions in dynamic environments. Analysts project the humanoid and mobile robotics market to reach $243 billion by 2035, climbing to more than $5 trillion by 2050, fueled by structural labor shortages, rising operational costs and the need for resilient, always-on production.

UMA's founders firmly believe this shift requires a new kind of robotic intelligence: one that is data-driven, adaptable, self-improving and safe enough to operate shoulder-to-shoulder with human teams.

Built by Engineers and Researchers Who Helped Define Modern AI

UMA's founding team represents decades of breakthroughs across the US, Europe and Asia: Remi Cadene pioneered Tesla Autopilot and Optimus, and democratized robot learning with LeRobot at Hugging Face, Pierre Sermanet advanced deep learning and robotics research for two decades at New York University and Google DeepMind, Simon Alibert co-founded LeRobot and brings expertise in scalable learning infrastructure, Robert Knight designed humanoid robots for 25+ years and open-sourced the widely used SO-100 robot.

Together, they bring deep technical expertise and a shared belief: the world needs robots that aren't just impressive in demos, but reliable in production.

Solving the Hardest Global Operational Challenges

Industrial and societal pressures are making advanced robotics not just desirable, but inevitable.

Across logistics, healthcare and aging societies, the constraints are structural:

-- In warehousing and logistics, labor is often the single biggest cost driver, representing up to 50% of total warehouse operating expenses, and annual warehouse worker turnover in the US frequently exceeds 40%, far above the national average (Business Plan Templates).

-- Globally, healthcare systems are heading toward a shortage of around 10 million health workers by 2030, including an estimated 4.8 million nurses and midwives, despite being the backbone of care delivery.

-- The world is aging fast: people aged 65 and over already account for about 10% of the global population and are projected to reach 16% by 2050, or roughly 1.6 billion people, reshaping demand for care, assistance and autonomy.

These are not abstract statistics. They determine whether hospitals stay staffed, whether supply chains keep moving, and whether older adults can live safely and independently.

UMA wants robotics to be a lever for resilience: machines that handle physical strain, repetitive tasks and environmental unpredictability, so people can focus on higher-value, human work.

Two Systems Designed for Complementary Real-World Use

UMA is developing two complementary systems.

1. A mobile industrial robot with dual arms designed for environments such as warehouses and assembly lines, where precision and repeatability are essential.

2. A compact humanoid robot designed to navigate human-centric spaces and collaborate directly with workers.

Together, these systems offer a pragmatic path toward high-reliability robotics deployment, balancing innovation with the real operational need of industry.

A Human-Centered Vision to Robotics

UMA's approach is intentionally grounded in durability, safety and real-world performance. The company builds systems that are lightweight, repairable, with a commitment to full civilian use. UMA believes robotics should enhance human capability; bringing more safety to people, time and new possibilities.

Built for Global Scale, Backed by Global Leaders

UMA is now hiring engineers, researchers and operational specialists from around the world who want to push the boundaries of physical intelligence. The company values high standards and firmly believes world-class innovation thrives on diverse perspectives and backgrounds.

UMA is supported by global investors such as Greycroft, Relentless, Unity Growth, >Commit, Factorial, ALM Ventures and Drysdale, along with leading figures in AI including Olivier Pomel, Yann LeCun, Thomas Wolf, Soumith Chintala and Nicolas Rosberg.

Sources (selection)

Gartner, Hype Cycle for Advanced Mobile Robots 2024 ; Gartner Press Releases 2024-2025 ; Data Bridge Market Research (Warehouse Robotics, Service Robotics, Cloud Robotics) ; United Nations World Urbanization Prospects ; European Commission Eurostat 2025 ; European Data Journalism Network ; EFN - European Federation of Nurses ; Le Monde ; Euronews ; MWPVL International ; Capstone Logistics ; INSEE 2022 ; DREES ; Ministry of Ecology ; Oxfam France ; ONU ; ADEME ; Sami Éco.

About UMA

Founded in Paris in 2025, UMA develops mobile and humanoid robots powered by advanced physical intelligence and designed for real-world performance. Its team brings experience from Tesla, Google DeepMind, Nvidia and Hugging Face. Backed by a significant first round, UMA is accelerating development and preparing pilot programs in logistics, manufacturing and healthcare in 2026. UMA's vision is to deliver intelligent robots that make work safer, that improve productivity, and make daily life better for millions.

View source version on businesswire.com: https://www.businesswire.com/news/home/20251201209844/en/

Contact

Press ContactsOutcast France - Strategic Communications Agency

© 2025 Business Wire, Inc.Disclaimer:This press release is not a document produced by AFP. AFP shall not bear responsibility for its content. In case you have any questions about this press release, please refer to the contact person/entity mentioned in the text of the press release.