Gasgoo Munich- On May 20, local time, Tesla wrapped up final deliveries of the Model S and Model X at its Fremont, California plant — and immediately began tearing the line down. Within four months, the facility will fully transform into a dedicated humanoid robot production line, targeting an initial annual capacity of 1 million units. In parallel, a second-generation line at the Texas plant is pushing forward, with long-term capacity sights set on 10 million.
Bolstered by a pilot production line already operational since November 2025 and a capital expenditure budget for 2026 raised to more than $25 billion, Tesla's humanoid robot is flooring the gas pedal. The project is shifting abruptly from the experimental R&D phase into a full-scale mass production lane.
Line Restructuring and Tech Reuse: The Robot as a "Smart Car with Legs"
On the surface, this looks like a simple reshuffling of production assets. But viewed through the lens of technology, Tesla's underlying logic is one of high reuse. As the company puts it internally: "A car is a robot on four wheels; a robot is a smart car with arms and legs."
Humanoid robots share battery modules, motor units, and core AI systems with their electric counterparts. The neural networks powering walking, perception, and manipulation are identical in origin to those behind Full Self-Driving (FSD).
The immediate payoff from this shared lineage is the cross-platform migration of training data. By May 2026, FSD had logged over 10 billion miles of real-world driving globally. The object recognition, path planning, and real-time decision-making capabilities distilled from these vast traffic scenarios can be directly ported to the robot's physical interactions. Conversely, models the robots learn regarding dynamic balance and fine manipulation can feed back into autonomous driving, creating a mutually reinforcing AI loop.
The logic holds on the manufacturing floor as well. The batteries, motors, electronic controls, and thermal management systems for robot production come almost entirely from the existing EV supply chain and production expertise. Tesla doesn't need to rebuild its industrial base; it is simply shifting its vehicle-making capabilities wholesale.
The Fremont line conversion is, in essence, a "reprogramming" of production resources — swapping the rhythm of car assembly for the rhythm of building "humans" within the same space.
The third-generation humanoid robot is set to debut and enter mass production in 2026. Unlike earlier prototypes, the new design targets industrial-grade reliability and aggressive cost competitiveness, aiming to replace repetitive and dangerous manual labor in homes, factories, and healthcare settings. With more than half of that $25 billion capital expenditure directed toward AI computing power and robotics manufacturing equipment, the scale of Tesla's wager on this new business is clear.
The 10-Billion-Unit Gamble: The Other Leg of an Abundant World
Musk's forecast for humanoid robots is radically aggressive: he puts long-term global demand at a potential 10 billion units. With the current global car parc at roughly 1.5 billion, 10 billion robots implies more than one for every person on Earth. In Tesla's narrative, robots are positioned as "labor replacement" infrastructure — as ubiquitous as home appliances and smartphones are today.
That figure aligns with the company mission updated in January 2026: "To build a world of extraordinary abundance." The new mission rests on two pillars: the Robotaxi network, designed to fundamentally overhaul the cost structure of transportation; and humanoid robots, intended to liberate humans from drudgery, repetition, and danger.
Musk has even suggested that as autonomous driving becomes ubiquitous, manual driving will gradually become a niche pursuit. Following that logic, once technology drives down the costs of transportation and physical labor, "universal abundance" ceases to be a fantasy.
Image source: Tesla
Yet, moving from blueprint to reality involves weighing more than a few heavy constraints.
The supply chain faces the pressure first. For core components like high-precision reducers, torque sensors, and dexterous hands, the global supply chain is nowhere near the scale capabilities of the automotive parts industry. Even with Tesla's formidable integration strength, supporting an annual capacity of 10 million units will require stretching the entire upstream ecosystem to its breaking point.
Cost is the second hurdle. Musk has consistently stressed that pricing must be low enough to drive adoption, yet unit costs for industrial-grade bipedal robots remain stubbornly high. Whether early mass production can drive costs down to a competitive level will dictate the speed of commercialization.
Real-world deployment is even trickier. Factories offer structured environments, but homes and healthcare settings are highly unstructured. Walking safely, grasping fragile objects, and interpreting vague instructions are far more complex than fixed-point assembly. The entire industry remains in the early stages of exploration — Tesla included.
Finally, social acceptance and regulatory frameworks are virtually blank slates. Introducing 10 billion robots into daily life will trigger concentrated outbreaks of issues regarding employment, privacy, and ethics. And the pace of global regulation is unlikely to catch up with the speed of technological advancement.
The Fremont conversion is viewed internally at Tesla as a "critical milestone." Musk's bet is that by leveraging the proven tech stack, manufacturing system, and data flywheel of its electric vehicle business, it can rapidly replicate the scale capabilities of an entirely new category.
The debut of the third-generation robot's mass production will be the first major test of this gamble. The actual output efficiency of the lines, the robots' performance in real-world scenarios, and early market feedback will provide the initial answer to whether a future of 10 billion units is a genuine wave or an overhyped narrative.
