AI • Technology • Digital InnovationIdeas, tools and guides for the future
AI GAME CHANGER PRO
Discover AI. Explore Technology. Create More.

This Humanoid Has No Legs — and It Just Hit 98% on Real Factory Work

November 20, 2026 • Ai Game Changer

Most humanoid robots chase the same dream: two legs, two arms, a general-purpose worker. Embodied AI, a Swiss startup, just went the other way — and the numbers are hard to ignore. Its first commercial humanoid, AIR, has no legs at all. It rolls on wheels, and it just hit 98% reliability on cable-handling tasks running on live factory production lines.

What is AIR?

Embodied AI was founded by three PhDs from EPFL, ETH Zurich and Cambridge. It unveiled AIR at Humanoid Europe 2026 on 15 November. The company says it went from research prototype to a factory-floor machine with signed OEM contracts in nine months.

AIR skips legs entirely. It moves on wheels and uses a tendon-driven arm design, with the motors and gearboxes housed in the torso rather than the arms. That keeps the arms light, and lets them yield on contact instead of fighting back — which matters on a factory floor where people and machines share the same aisle.

The robot is built to ISO/TS 15066:2016, the standard covering collaborative robots that work without safety cages. Embodied AI plans CE marking in 2027, which it needs to sell AIR more broadly across the European market.

How it learns a new task

Instead of training a model offline and deploying it cold, Embodied AI has human operators guide the robot through a task directly on the production line. The system is designed to pick up the skill from that hands-on demonstration within roughly a month, after which it runs with less supervision.

The result, according to the company: 98% reliability on cable-handling — above the roughly 90% threshold it says manufacturers generally require before trusting a task to automation.

That approach is a direct answer to a problem that has slowed other humanoid programs. Models trained in simulation or on borrowed datasets often stumble when they meet a specific factory's parts, lighting and layout. By learning on the actual line from day one, Embodied AI is betting it can close that gap faster than competitors relying on large, general-purpose foundation models.

Co-founder and CEO Francesco Stella put the philosophy bluntly: "Hallucination of AI models is acceptable in the digital space. In robotics, however, incorrect actions have immediate and irreversible physical consequences." That is why AIR's design is about risk tolerance, not ambition — a narrower, safer machine for a narrower set of jobs.

Why this matters

The pitch lands at a moment when manufacturers are short-staffed and humanoid robotics has struggled to move from demo reels to repeatable production work. Embodied AI cites projections of as many as 1.9 million unfilled US manufacturing jobs by 2033 as the backdrop for its bet on collaborative, wheel-based humanoids that don't need a redesigned facility to operate.

AIR also arrives as rivals pursue very different paths to the same factory floor. Apptronik's Apollo 2 recently posted 96% accuracy sorting parts on a live Mercedes-Benz line using Google's Gemini Robotics 2.5 model, and GXO and Apptronik are planning trials toward a possible 1,000-unit deployment. Chinese manufacturers including AgiBot and UBTech are scaling bipedal humanoids at high unit volumes.

Embodied AI's bet is narrower: faster time-to-deployment on a smaller set of manual tasks, rather than a general-purpose bipedal worker. A wheeled robot that learns one factory's cable job in a month and does it at 98% is, for a manufacturer, a much easier yes than a walking robot that promises everything.

What happens next

Embodied AI's Series A was led by Faber, with TechShop, Look AI Ventures, Plug and Play and Vento also participating; the company has not disclosed the round's size. The near-term plan is to expand signed OEM contracts beyond cable-handling into other manual assembly tasks, while working toward the 2027 CE marking.

One honest note: the 98% figure is the company's own report from its own production lines, not an independent test. But the direction of travel is clear — the humanoid race is splitting into two camps, and the wheeled, learn-on-the-line camp just posted a number the bipedal camp has to beat.

Bottom line

AIR proves a point the industry needed to hear: a humanoid does not need legs to be useful. Wheels, tendon-driven arms, no safety cages, and a month of learning on the real line got it to 98% on real factory work. Watch whether the 2027 CE marking and the next round of OEM contracts turn a promising demo into a product manufacturers actually buy.

Source: The Robotics Media — Embodied AI's AIR Humanoid Hits 98% on Factory Cable Tasks. Hero photo: robotic arm polishing guitars at the Martin Guitar Factory, by Henrysz, CC BY 4.0, via Wikimedia Commons.