Tesla has started building its Optimus humanoid robot at the Fremont, California factory, repurposing floor space that used to turn out electric vehicles. The company confirmed the move on Wednesday, October 21, 2026, pairing the manufacturing news with a Q3 earnings report that showed net income down 43% year over year. The timing was not an accident: Tesla is leaning on Optimus as its next growth story at the exact moment its core car business is cooling off.
The announcement, first reported by TechCrunch, lands as investors try to figure out whether Tesla is still primarily a car company or something closer to a robotics and AI firm that happens to also sell cars. The numbers released alongside the Optimus update make that question harder to dodge. Third-quarter net income came in at $787 million, down sharply from $1.37 billion in the same quarter of 2025, even as the company pushes hard into humanoid manufacturing it hopes will eventually dwarf the automotive business.
Don't miss new tech stories on Google
Add Tech Insider once in the Google app and our stories appear in your news suggestions.
What Tesla Actually Announced on October 21
Tesla said the robots now coming off its Fremont line are the company’s first “production-quality” Optimus units, a notable shift in language from earlier updates that described pilot builds and small test batches. The company also said its current manufacturing process is “suitable for a commercial product,” a specific phrase chosen to signal that Optimus has moved past the prototype stage and into something resembling a real assembly line.
What Tesla did not say matters just as much. The company has not released a specific production volume for the new line, has not given a delivery timeline for outside customers, and has not published a commercial price. There is no confirmed robot count moving through Fremont each week under this new process, and no sale date for buyers outside Tesla’s own operations. Anyone searching for a release date or sticker price for Optimus right now will not find one in this announcement; Tesla has kept those details closed for now.
The space itself is symbolically loaded. Fremont is the plant that built the Model S starting in 2012 and has served as Tesla’s flagship US assembly site ever since. Converting part of that floor to humanoid robot production, rather than building a dedicated greenfield robotics plant, tells its own story: Tesla is treating Optimus as a direct extension of its automotive manufacturing know-how, not a separate venture bolted on from outside.
The Earnings Numbers Behind the Robot News
Tesla’s third-quarter 2026 results, covered in detail in our prior Q3 earnings breakdown, showed revenue that was largely flat against the prior year and expenses that climbed. Net income fell 43% to $787 million from $1.37 billion a year earlier. Tesla attributed part of the gap to higher spending, without providing extra detail on the earnings call framing that this reporting cycle covers.
That profit decline arrives even though Tesla had already flagged, in our earnings preview coverage, that Optimus spending and AI infrastructure costs were likely to weigh on margins this quarter. The pattern fits a company making a deliberate bet: spend aggressively now on a new product line, accept a hit to near-term profit, and wait for Optimus to become a second major revenue stream alongside vehicles and energy storage.
| Metric | Q3 2026 | Q3 2025 | Change |
|---|---|---|---|
| Net income (GAAP) | $787 million | $1.37 billion | Down 43% |
| Revenue | Largely flat | Baseline | Roughly unchanged |
| Operating expenses | Higher | Baseline | Increased |
| Optimus production status | Production-quality units, Fremont line | Pilot/test builds | Advanced to commercial-grade process |
Tesla framed the Optimus manufacturing milestone as evidence that the investment is paying off in engineering terms, even if it has not yet shown up as new revenue. The company’s own wording, that the process is “suitable for a commercial product,” is doing a lot of work here: it is a claim about manufacturing readiness, not about near-term sales or profitability from the robot line itself.
Why Tesla Is Betting on a Humanoid Robot Right Now
The logic behind Optimus has been consistent since Tesla first previewed the robot: Elon Musk has argued that a general-purpose humanoid machine, capable of working factory floors, warehouses, and eventually homes, represents a market larger than the entire auto industry. Whether or not that framing holds up, Wednesday’s announcement shows Tesla treating Optimus as a near-term manufacturing priority rather than a long-range research project.
That shift tracks with what we covered back when Tesla said it was building roughly 1,000 Optimus units a week, a cadence that, if sustained and scaled, would put real pressure on humanoid robotics rivals still operating at much smaller volumes. Moving that output onto a converted EV line at Fremont, rather than a smaller pilot facility, suggests Tesla wants Optimus manufacturing to inherit the same high-volume discipline that made the Model 3 and Model Y profitable at scale.
There is also a defensive angle. Tesla’s core EV business has faced slowing growth in several markets, and the company has repeatedly pointed to AI, robotics, and autonomy as the next leg of its valuation story. Optimus production starting now, even without confirmed volume or pricing, gives Tesla a concrete milestone to point to heading into a period when vehicle margins are under pressure.
What Tesla Executives Said About the Ramp
On the earnings call tied to this report, Tesla CEO Elon Musk described the current stage bluntly. “We’ve started what I’d call early production on our high-volume line in Fremont,” Musk said, according to a report on the call. He was careful to temper expectations about how smoothly that ramp would go, adding that “the ramp would be somewhat bumpy because it is such a radically new product.”
Musk went further in laying out just how fragile early manufacturing ramps can be for a product this complex. “There’s like 10,000 things that have to go right for production of the robot, and it’ll move as fast as the slowest, least lucky of the 10,000,” he said. It is a line that doubles as an explanation for why Tesla has avoided committing to a hard production number or delivery date even while announcing that manufacturing has begun.
Pressed on when the bottlenecks might clear, Musk pointed to a timeline measured in years rather than months: “I do expect these issues to be solved around the end of next year.” That comment suggests Tesla itself views 2026 production as the opening phase of a ramp that will not reach full-volume, fully reliable output until sometime in 2027.
Tesla’s own shareholder materials echoed the same cautious-but-confident tone in writing, stating that “while we will continue to iterate on the hardware and manufacturing process, we have proven that the current manufacturing process for Optimus is suitable for a commercial product.” Together, the spoken and written comments paint a picture of a company that believes it has cleared a key engineering hurdle while openly acknowledging the harder work of scaling is still ahead.
How Optimus Compares to Other Humanoid Robot Projects
Tesla is not the only company racing to put a humanoid robot into real-world or industrial use. Figure AI has been developing general-purpose humanoid robots aimed at industrial and commercial tasks and is widely viewed as one of the closest direct competitors to Optimus in ambition, if not yet in manufacturing scale. Agility Robotics has focused its Digit robot on warehouse logistics work rather than general-purpose tasks, giving it a narrower but more immediately commercial niche. Boston Dynamics continues advanced research work on its Atlas platform alongside its established industrial robotics line, while Unitree has pushed into lower-cost humanoid hardware aimed at a different price tier entirely.
What sets Tesla apart in this race is not necessarily the robot’s capabilities, which remain unverified at the production-volume level, but the manufacturing infrastructure behind it. Few humanoid robotics companies have access to an automotive-scale factory with a decade of high-volume assembly experience to repurpose. That is the real asset Tesla brought to Wednesday’s announcement: not a new robot capability, but a credible claim that it can build humanoid machines the same way it builds cars, at a pace most competitors cannot easily match.
| Company | Primary Focus | Manufacturing Scale | Target Market |
|---|---|---|---|
| Tesla (Optimus) | General-purpose humanoid | Repurposed automotive line, Fremont | Internal use, eventual commercial sale |
| Figure AI | General-purpose humanoid | Smaller-scale, not auto-plant level | Industrial and commercial work |
| Agility Robotics | Warehouse logistics (Digit) | Focused deployment, not mass production | Logistics and fulfillment centers |
| Boston Dynamics | Research platform (Atlas) plus industrial robotics | R&D-driven, limited unit volume | Research, select industrial clients |
| Unitree | Lower-cost humanoid hardware | Growing but smaller than auto-scale | Price-sensitive buyers, research labs |
None of these competitors has published a confirmed September or October 2026 production milestone comparable to what Tesla just announced, which is itself informative. It suggests Tesla is, for now, the company most willing to publicly tie a specific manufacturing claim to a specific date, even while withholding the volume and pricing numbers that would let outsiders judge how real that claim is at scale.
The Market Reaction and What Wall Street Is Weighing
Investors had two conflicting data points to digest on the same day: a 43% profit decline and a headline manufacturing milestone for a product Tesla has promoted as its long-term growth engine. The company’s revenue came in largely flat, and expenses rose, which on its own would typically be read as a disappointing quarter for an automaker. Layered on top of that is the Optimus news, which gives bulls a reason to look past the near-term numbers toward Tesla’s robotics and AI ambitions.
This is not a new pattern for Tesla. The stock has repeatedly traded less on quarterly automotive profitability and more on the market’s belief in Musk’s longer-range bets, whether that is full self-driving, robotaxi plans, or now Optimus. A 43% drop in net income would be a serious red flag for a traditional automaker’s stock price; for Tesla, it is being weighed against a robot manufacturing claim that has no confirmed revenue attached to it yet.
That dynamic also explains why Tesla chose to announce the Optimus production milestone in the same breath as its earnings, rather than as a standalone update. Pairing a weak profit number with a forward-looking manufacturing story is a familiar investor-relations tactic, and it puts pressure on analysts to frame the quarter around Optimus rather than around the margin compression in the core auto business.
Historical Context: From Robot Suit Reveal to Factory Floor
Tesla first teased Optimus at its 2021 AI Day event, where the robot appeared as little more than a concept, famously represented on stage by a dancer in a robot suit. The gap between that reveal and Wednesday’s claim of production-quality units rolling off a converted car assembly line spans roughly five years, a timeline that mirrors how long it has taken other ambitious Tesla projects, like the Cybertruck and full self-driving, to move from announcement to anything resembling commercial reality.
Along the way, Tesla has used its Fremont factory as a proving ground for manufacturing claims before, not always smoothly. Musk once described Model 3 production as “manufacturing hell” before the line eventually scaled into one of the highest-volume EV programs in the US. The Optimus ramp, based on Musk’s own description of “10,000 things” needing to go right, suggests Tesla expects a similarly bumpy path this time, just with a humanoid robot instead of a sedan.
Impact on Tesla’s Broader AI and Robotics Strategy
Optimus does not exist in isolation inside Tesla’s roadmap. It sits alongside the company’s self-driving software stack, its custom AI chip efforts, and its broader push to position itself as an AI company rather than strictly an automaker. The chip side of that strategy has not been friction-free; our earlier reporting on Tesla’s dispute with Samsung over AI5 and AI6 chip supply showed how dependent Optimus and other AI-heavy projects are on external silicon partners staying aligned with Tesla’s production timelines.
That dependency matters because a humanoid robot built at automotive scale needs a steady, high-volume supply of the compute hardware that lets it perceive and navigate its surroundings. Any slippage in chip supply would directly affect how fast Tesla can turn Wednesday’s production announcement into the kind of weekly output numbers the company has previously discussed. It is also a reminder that Optimus’s success depends on more than Tesla’s own assembly line discipline; it depends on a chip supply chain that has already shown signs of strain.
More broadly, prediction markets have already started pricing in how fast humanoid robots might scale across the industry as a whole, not just at Tesla. Betting activity tracked in our coverage of Manifold’s humanoid robot volume odds shows traders assigning high confidence to tens of thousands of humanoid units shipping industry-wide before the end of the decade, a bar that Tesla’s Fremont announcement alone does not yet confirm it can clear, given the absence of a published unit count.
The Funding and Competitive Landscape Around Robot AI
Tesla’s move also lands in a market where capital is actively chasing robot-focused AI software, not just hardware. Smaller players have raised meaningful funding rounds specifically to build the software brains that could run on humanoid or home robots, as seen in our reporting on Dream Labs’ $90 million seed round for robot AI. That kind of investor appetite suggests the market sees humanoid and general-purpose robotics as a genuine emerging category, not just a Tesla-specific side project.
For Tesla, that broader competitive pressure is a double-edged sword. On one hand, outside investment validates the thesis that humanoid robots are a real market worth building toward. On the other, it means Tesla cannot assume it will have the field to itself by the time Optimus reaches genuine commercial volume, whenever that turns out to be, given Musk’s own acknowledgment that the harder manufacturing problems may not be fully solved until around the end of 2027.
What a 43% Profit Drop Signals About Tesla’s Core Business
It is worth separating the Optimus story from what the earnings number says about Tesla’s traditional car business. A 43% year-over-year decline in net income, against revenue that stayed largely flat, points to margin compression rather than a collapse in demand. Higher expenses, some tied to AI and robotics investment, are eating into profit even as the top line holds roughly steady.
That is a meaningfully different story than a demand-driven slowdown, and it is one Tesla appears comfortable telling, given that it chose to pair the weaker profit figure with the Optimus milestone rather than downplay it. The message to investors seems to be that near-term margin pain is the deliberate cost of building toward a much larger robotics business, not a sign that the core EV franchise is in trouble.
Predictions: Where Optimus Production Goes From Here
- Tesla will likely disclose a specific weekly or monthly Optimus unit count within the next one to two quarters, once the Fremont line stabilizes enough to generate a number worth publicizing.
- Expect continued margin pressure through at least the next few quarters as Optimus and AI infrastructure spending stays elevated relative to the scale of any resulting revenue.
- Given Musk’s own comment about problems being resolved “around the end of next year,” a meaningful commercial pricing and delivery announcement for outside buyers is unlikely before late 2027.
- Rival humanoid robotics firms, including Figure AI and Agility Robotics, will face growing pressure to publish their own manufacturing-scale milestones to avoid ceding the public narrative to Tesla.
- Tesla’s chip supply relationships, including the Samsung AI5 and AI6 arrangement, will remain a recurring bottleneck risk that could directly slow any Optimus production ramp if not resolved cleanly.
What This Means for Tech Workers and Investors
For software engineers and AI practitioners watching from outside Tesla, Wednesday’s announcement is a signal that robotics manufacturing, not just robotics software, is becoming a serious hiring and investment category. Companies that can combine machine learning expertise with actual factory-floor production experience, the exact combination Tesla is trying to leverage at Fremont, are likely to have an edge in a market where plenty of firms can build a robot demo but far fewer can build thousands of them reliably.
For investors, the quarter is a reminder that Tesla’s stock story increasingly depends on how convincingly the company can translate manufacturing claims into verified production numbers. A 43% profit decline paired with an unquantified robot announcement is, by itself, more of a promise than a result. The next few earnings cycles, where Tesla will either produce a specific Optimus unit count or continue speaking only in general terms, will likely determine how much credit the market gives this story going forward.
Frequently Asked Questions
Did Tesla announce a release date or price for Optimus?
No. Tesla confirmed that production-quality units are being built at Fremont, but the company has not published a commercial price or a delivery date for outside buyers.
How much did Tesla’s profit fall in Q3 2026?
Net income fell 43% year over year, to $787 million from $1.37 billion in the same quarter of 2025, according to Tesla’s own reporting.
Where is Tesla building Optimus robots?
Production is happening at Tesla’s Fremont, California factory, in manufacturing space that was previously used for electric vehicle production.
What does “production-quality” mean for Optimus?
Tesla says these are the first units built on a high-volume line using a manufacturing process the company considers suitable for a commercial product, as opposed to earlier pilot or prototype builds.
Who are Tesla’s main competitors in humanoid robotics?
Figure AI, Agility Robotics, Boston Dynamics, and Unitree are the most commonly cited competitors, though none has published a comparable production-scale milestone as of this report.
Why did Tesla’s profit drop while announcing a major manufacturing milestone?
Tesla reported largely flat revenue alongside higher expenses, including continued investment in AI and Optimus, which weighed on net income even as the company highlighted progress on robot manufacturing.
When will Tesla sell Optimus to customers outside the company?
Tesla has not given a firm date. Comments attributed to Musk on the earnings call suggest key production issues may not be resolved until around the end of 2027.
Is Optimus replacing Tesla’s car manufacturing at Fremont?
No. The announcement describes Optimus occupying space previously used for EV production, not a full replacement of Tesla’s vehicle manufacturing at the site.
