A Humanoid Robot Just Clocked In at Mercedes-Benz — and Hit 96% on the Line
For years, humanoid robots have looked great in lab demos and viral clips. This month, one of them clocked in somewhere far less forgiving: a live car factory. On November 10, 2026, Google DeepMind introduced Gemini Robotics 2.5, its newest AI model for controlling robots — and alongside it came a headline number. In a proof of concept with Apptronik and Mercedes-Benz, Apollo 2 humanoid robots reached a 96% success rate on real production lines. Here is what actually happened, why it matters, and why the remaining 4% may be the most interesting part of the story.
The 96% Test: A Real Factory, Not a Lab
The test did not happen on a stage. It ran on active Mercedes-Benz production lines in Kecskemét, Hungary, and Sindelfingen, Germany — working factories, not staged sets. The partners were Google DeepMind, which supplied the AI; Apptronik, the Austin-based maker of the Apollo 2 humanoid; and Mercedes-Benz, which provided the factory floor. Mercedes-Benz is also an investor in Apptronik.
The job was “kitting.” The robot picks metal brackets out of bins filled at random and places them into sequencing trays, in the exact order each car coming down the line needs them. It sounds boring, and that is exactly why it matters. Kitting is repetitive, physically tiring, and messy, because parts never sit in the same position twice. Classic factory robots hate that kind of mess — they need parts presented perfectly, which means expensive fixtures and re-engineering every time something changes.
Two AI Brains in One Body
The interesting technical idea is that the work is split between two models. Gemini Robotics ER acts as the planner: it tracks the multi-step workflow, watches progress, and works out how to recover when something goes wrong. Gemini Robotics 2.5 is the hands: a vision-language-action model that turns what the robot sees into precise arm and finger movements. Think of it as a supervisor and a skilled worker sharing one body.
Google says the new model completes tasks faster than its predecessor and shows big jumps in success across more than 200 tasks, with some tasks going from almost never working to working most of the time. Apptronik says the models improved dramatically within a few weeks of extra training and carried over to the Mercedes-Benz lines with very little drop-off.
Why the Car Industry Is the Proving Ground
Car makers run some of the most demanding production lines in the world. If humanoids can earn a place there, warehouses, logistics, and electronics assembly are next in line. There are two deeper reasons the factory floor is where this technology will be won or lost.
First, robots that adapt instead of being programmed. Traditional industrial robots follow exact scripts. An AI-driven humanoid can, in principle, be retrained for a new task with data and practice instead of months of re-engineering. That is the real promise of “physical AI.”
Second, human-shaped robots fit human-shaped factories. Factories are built for people: aisles, shelves, bins, and workstations at human height. A humanoid can slot into an existing line without rebuilding it.
The 4% Problem Nobody Is Talking About
Ninety-six percent sounds amazing until you think like a factory manager. A production line cannot pause. If one pick in 25 fails, someone has to notice and fix it, and a missing bracket can hold up a car. Human workers doing the same job are close to error-free. So 96% is a huge step for robots, but probably not yet good enough for full production.
There are other honest caveats. DeepMind itself calls this a proof of concept, not a fleet running at scale. Key numbers have not been published: how many attempts were made, how often a human had to step in, cycle times, maintenance, and cost compared with conventional automation. And Gemini Robotics 2.5 is rolling out to selected robotics partners only, including Apptronik, Agile Robots, and Boston Dynamics. There is no public access yet.
Still, this test was much tougher than a lab demo: real bins, real parts, real sequencing, and the everyday chaos of a working factory.
Safety Moves to the Front of the Line
Alongside the model, DeepMind released an open safety benchmark called ASIMOV-Jailbreak. It tests whether robot AI can be tricked into unsafe physical actions — the robot equivalent of “jailbreaking” a chatbot. As AI moves from screens into machines that can lift and move things, this kind of testing matters a lot more.
The Wider Race Is Speeding Up
The Mercedes-Benz test is not happening in isolation. The whole humanoid field is shifting from crowd-pleasing stunts toward machines that can handle tools, move goods, and complete tasks with less human intervention — and next year’s CES in Las Vegas, running January 6 to 9, will be the next big test of that shift.
Chinese companies accounted for 21 of the 38 humanoid exhibitors at CES 2026, although Hyundai Motor Group-controlled Boston Dynamics took CNET Group’s Best Robot award with Atlas. Unitree Robotics and AgiBot each shipped more than 5,000 humanoids in 2025, out of approximately 15,000 worldwide, according to figures reported by The Associated Press.
The Korean giants are pushing hard too. LG Electronics plans to unveil a bipedal humanoid in the first quarter of 2027 using Nvidia’s Isaac GR00T model and Jetson Thor computing platform, and it signed an agreement with the Korea Institute for Robot Industry Advancement on November 13 to build humanoid safety assessment standards. Hyundai Motor Group aims to establish annual production capacity for 30,000 Atlas robots by 2028 as it prepares to introduce the machines into automotive manufacturing. Tesla said in October that early production of its Optimus humanoid had begun at its Fremont factory. Industry sources say Samsung plans to unveil two humanoids at CES 2027, although Samsung has not confirmed the plans.
Park Dong-il, head of the advanced robotics research center at the Korea Institute of Machinery and Materials, put it plainly: “More than half will probably move toward autonomous work.” And You Bum-jae, a principal research scientist at the Korea Institute of Science and Technology, offered a sober benchmark: “Task success rates for models around the world are still below 90 percent. To unveil physical AI, performance has to go beyond that.” The Mercedes-Benz result, at 96%, is the first headline number to clear that bar on a real production line.
The Bottom Line
For years, humanoid robots were impressive demos looking for a job. This is one of the clearest signs yet that they are starting to find one. The 96% figure is not the finish line, but it is the first time a general-purpose humanoid has shown reliable results on a real production line. Watch for three things next: whether the success rate climbs toward 99% and beyond, whether Mercedes-Benz expands the pilot beyond these two plants, and whether other car makers follow. The robots have clocked in. Now they have to prove they can stay on the line.
Sources: Gardaran, “Humanoid Robots on a Mercedes-Benz Line: Gemini Robotics 2.5 Hits 96%” (November 19, 2026); Aju Press, “CES 2027 humanoids face a work test” by Kim Dong-young (November 20, 2026); Google DeepMind (via Gardaran); The Associated Press (via Aju Press).
Edited by Muse @QUE.COM
Website: https://QUE.COM Intelligence
Sponsored by: https://MAJ.COM AI Autonomous
Never miss a customer call again — MAJ.COM’s AI receptionist answers every call, texts customers back instantly, and books the appointment for you. See how it works at https://MAJ.COM
Discover more from QUE.com
Subscribe to get the latest posts sent to your email.
