Tesla has put a real date on its humanoid robot ambitions. On the company’s third-quarter 2026 earnings call, held Wednesday, October 21, CEO Elon Musk said Optimus deliveries to companies outside Tesla are expected to begin around the end of 2027, as the automaker confirmed it has started early production of the robot on a high-volume line at its Fremont, California factory. The announcement, folded into a shareholder update that also detailed Tesla’s broader financial results, gives investors and industry watchers their first concrete commercial timeline for a project Musk has described for years as the company’s largest long-term opportunity.
The headline framing making the rounds, that Tesla is aiming to begin humanoid robot sales “next year,” checks out against Musk’s own words, since the Q3 2026 call took place in October 2026, meaning end-of-2027 deliveries genuinely fall within the next twelve-plus months from today, October 22, 2026. But the specifics behind that framing, how production actually started, what still needs fixing, and why 20,000 units a week is the real number to watch, are more complicated than a single headline can capture. Here is what Tesla has confirmed, what remains unverified, and what it means for the humanoid robotics race heading into 2027.
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 Confirmed on the Q3 2026 Earnings Call
Tesla’s Q3 2026 shareholder deck and the subsequent earnings call laid out the Optimus program’s status in more detail than previous quarters. According to the earnings call transcript as reported by Benzinga, Tesla said it has begun early production of Optimus on a high-volume manufacturing line in Fremont, the same facility that once built electric vehicles exclusively. Musk told analysts the line has “the vast majority” of equipment installed for a production system designed around an annualized rate of roughly one million units per year.
Musk was direct about the timeline for outside sales. “I do expect these issues to be solved around the end of next year. So that’s the time to expect when we’re delivering Optimus to companies outside of Tesla,” he told investors on the call, a remark first reported by Electric Vehicles. He also described the current manufacturing phase plainly: “We’ve started what I’d call early production on our high-volume line in Fremont.”
That is a meaningfully more specific statement than Tesla has offered in prior quarters, when Optimus updates leaned on demo videos and factory-floor appearances rather than a stated delivery window. The production start itself drew attention partly because it landed in the same earnings report where Tesla’s quarterly profit fell sharply, a detail Yahoo Finance highlighted in its coverage of the announcement. Pairing a robot milestone with a profit decline is exactly the kind of split narrative that tends to divide how Wall Street and robotics watchers read the same quarter.
Inside Fremont: How Early Production Actually Works
“Early production” is doing a lot of work in Tesla’s messaging, and it is worth unpacking. Tesla has said the Fremont units represent the first Optimus robots built to what the company calls a production-quality standard, according to reporting cited in the Q3 2026 materials. That is distinct from the prototype and pilot-line robots Tesla has shown off at prior events and in earlier factory tours. The company has also said its current manufacturing process is suitable for a commercial product, even though both the hardware and the process are expected to keep changing.
Musk framed the ramp-up as deliberately slow. He described Optimus’s growth curve as an S-curve, telling investors, “The growth of Optimus is an S curve, but the initial curve is a shallow slope,” according to commentary summarized by TradingView’s coverage of the call. In practical terms, that means Tesla expects output to climb gradually at first before accelerating, a pattern common in novel hardware categories where manufacturing processes, supply chains, and component sourcing are still being worked out simultaneously with the product design itself.
This mirrors a pattern tech-insider.org has tracked before with Optimus production milestones, where incremental weekly output figures have served as checkpoints toward Tesla’s much larger annualized targets. The difference this quarter is that Tesla, for the first time, tied a specific calendar year, 2027, to the point when outside companies would actually receive units, rather than describing only internal deployment or further prototype iterations.
The 20,000-Units-a-Week Math
Tesla’s stated production target for Optimus is 20,000 units a week, a figure that would translate to roughly 1.04 million robots a year if sustained continuously. Musk was careful to separate that target from the near-term delivery timeline, telling investors, “I would expect that it takes us probably at least a year, maybe a little longer, to get to that annualized rate, to get to 20,000 units a week,” as reported by Drive Tesla Canada’s account of the earnings call.
That single sentence actually contains two separate commitments that are easy to conflate. The first is the end-of-2027 target for beginning deliveries to outside companies at all, likely in limited volume. The second is the roughly one-year-or-longer runway to reach the full 20,000-per-week manufacturing rate, a ramp that would logically extend into 2028 or later if it starts from the current early-production baseline. In other words, Tesla beginning to sell Optimus to outside buyers in 2027 does not mean Tesla will be building a million robots a year by 2027. Those are different milestones on different clocks, and conflating them is the most common way this story gets oversimplified.
| Milestone | Status as of Oct. 22, 2026 | Source |
|---|---|---|
| Early production start (Fremont) | Confirmed, underway | Tesla Q3 2026 earnings materials |
| Production-line equipment installed | “Vast majority” installed for ~1M units/year line | Benzinga earnings call transcript |
| First deliveries to outside companies | Targeted around end of 2027 | MarketWatch live coverage |
| 20,000 units/week production rate | At least one year out, possibly longer, from Q3 2026 | Drive Tesla Canada |
| AI5 chip integration into Optimus | Discussed on call with an approximate future timing, contingent on Tesla’s plans proceeding as expected | Benzinga earnings call transcript |
| Confirmed retail or enterprise price | Not disclosed in Q3 2026 materials | Tesla Q3 2026 shareholder update |
| Named external customer or order | Not disclosed in Q3 2026 materials | Tesla Q3 2026 shareholder update |
Why “Next Year” Really Does Mean Late 2027
Some of the confusion swirling around this story stems from the phrase “next year” itself. Musk’s own comment on the call used that exact phrasing, saying the outstanding issues should be “solved around the end of next year.” Because the earnings call happened in October 2026, “next year” from that reference point is 2027, not 2026 and not 2028. That makes the widely circulated framing, that Tesla is aiming to begin humanoid robot sales next year, accurate as stated on October 22, 2026, provided readers understand “next year” means the calendar year 2027, with deliveries clustering toward its end rather than its start.
It is a subtle distinction, but an important one for anyone tracking the stock or the broader robotics market. A commitment to begin deliveries “by the end of 2027” leaves roughly 14 months of runway from today, not a matter of weeks. Businessinsider.com’s Africa edition, covering the same call, described the target explicitly as “shipping Optimus to customers by the end of 2027,” language that lines up with Musk’s own wording and removes any ambiguity about which year is actually meant.
What’s Still Unresolved: Design, AI, and Reliability
Musk did not present the 2027 target as a sure thing. He was explicit that Optimus’s hardware and software still need work before Tesla is comfortable shipping units to customers who are not Tesla itself. “There’s still some things we need to tweak about the design, because you need to deliver a robot that is safe and reliable and has long duration,” Musk said, a line that gets at the core engineering challenge: a humanoid robot destined for industrial, dangerous, repetitive, or around-the-clock work has to survive sustained use in ways a factory demo unit does not need to.
On the Q3 2026 call, Tesla management grouped the remaining work into four buckets: design, artificial intelligence, reliability, and operating life, all of which the company expects to largely resolve around the end of 2027, the same window attached to the first outside deliveries. That is not a coincidence. Tesla appears to be treating “ready to design around new constraints” and “ready to sell” as the same gate, which is a more conservative posture than simply picking a ship date and working backward from it.
Durability in particular is a recurring theme in how Musk talks about Optimus. In comments reported separately from the formal call, Musk has emphasized that a commercial robot “can’t, like, wear out after a month,” underscoring that Tesla sees longevity, not just capability, as the bar it needs to clear before shipping units to paying customers, a framing echoed in Business Insider’s coverage of the production announcement.
The AI5 Chip Question
One detail from the earnings call that has drawn less attention than the delivery date, but matters just as much for the robot’s actual capability, is silicon. Tesla’s Q3 2026 earnings call transcript indicated the company plans to integrate its AI5 chip into Optimus, with an approximate timing discussed on the call, subject to Tesla’s broader chip plans proceeding as expected. AI5 is the chip generation Tesla has also discussed in the context of its vehicle self-driving computer, and its path into Optimus has not been free of friction.
Tesla’s AI chip roadmap has been a point of tension with its manufacturing partners, something previously reported in detail regarding Tesla’s dealings with Samsung over AI5 and AI6 chip production. Whatever the outcome of that broader dispute, the Optimus delivery timeline is now at least loosely coupled to it: a humanoid robot capable of safe, reliable, long-duration operation in industrial settings needs a compute platform that can keep up, and AI5’s own schedule is one more moving part inside an already multi-variable timeline.
No Price Tag, No Preorders Yet
Despite the new delivery timeline, Tesla’s Q3 2026 shareholder update did not specify a customer, a commercial price, or an exact initial sales volume for Optimus. Reporting on the call is consistent on this point: no confirmed retail or enterprise price has been published, and there is no open public order process for Optimus as of this writing. Readers may encounter a $20,000 to $30,000 figure circulating online in connection with Optimus; that number reflects a prior target Musk has floated in past commentary, not a price Tesla has published or confirmed in its Q3 2026 materials.
Similarly, nothing in the current reporting indicates whether the first units shipped under the 2027 target will go to external, arms-length customers or to controlled deployments and pilot partners that Tesla manages closely, at least initially. Given that the stated early use cases are industrial, dangerous, repetitive, and around-the-clock work, a cautious, partner-led rollout before broader commercial availability would be consistent with how Tesla has typically introduced genuinely new hardware categories.
Musk’s Robot Timeline Track Record
Context matters here, and the context is that Tesla has a long, well-documented history of setting ambitious hardware and software timelines that later slip, sometimes by months, sometimes by years. That pattern is not unique to Optimus. It is the backdrop against which every Tesla delivery date, from Full Self-Driving capability claims to new vehicle programs, gets evaluated by analysts, journalists, and shareholders alike. Musk himself has effectively built that skepticism into this quarter’s messaging by repeatedly qualifying the 20,000-units-a-week target with phrases like “at least a year, maybe a little longer,” rather than committing to a hard date.
That hedging is itself informative. Compared with past Optimus updates, which leaned heavily on demonstration videos of robots performing choreographed tasks, this quarter’s language is noticeably more grounded in manufacturing specifics, a line count, a factory location, an equipment-installation status. Whether that added specificity translates into the end-of-2027 date actually holding is, naturally, unverifiable today. What can be said with confidence is that Tesla has now put a public, calendar-anchored stake in the ground in a way it had not done as explicitly in prior quarters, giving the market a concrete date to hold the company to.
How Optimus Fits Into the Wider Humanoid Robotics Race
Tesla is far from the only company chasing commercial humanoid robots. Figure AI, Boston Dynamics, Agility Robotics, 1X Technologies, and China’s Unitree are all developing humanoid platforms aimed at industrial, logistics, or home-assistance use cases, and the category as a whole has drawn heavy investor interest over the past two years. What stands out about Tesla’s Q3 2026 disclosure is less a specific competitive comparison, since none of those rivals has published matching production-rate or delivery-date figures in the days surrounding Tesla’s own announcement, and more the scale of ambition Tesla is describing: a factory line sized for roughly a million units a year is an order of magnitude beyond the pilot-scale manufacturing most of the humanoid robotics field has discussed publicly to date.
That scale is also why Optimus keeps getting discussed as a manufacturing story as much as a robotics story. Vertical integration, Tesla building its own line rather than outsourcing assembly, is the same playbook the company used for vehicles and for its energy storage hardware. The approach carries real execution risk, since Tesla is simultaneously trying to perfect the robot’s design, its AI stack, and its manufacturing process at once, but it also means Tesla controls more of its own timeline than a company relying on contract manufacturers would.
The broader robotics funding environment adds another data point. Venture investment into humanoid and industrial robot AI has continued to flow into the sector, including deals like the one covered recently involving Dream Labs, a reminder that Tesla is operating inside a well-capitalized, increasingly crowded field rather than a market it can simply define on its own terms.
Confirmed vs. Unconfirmed: Separating Fact From Speculation
Given how much speculation tends to attach itself to any Optimus headline, it is worth laying out precisely what Tesla has confirmed on the record versus what remains speculative, borrowed from past commentary, or simply unknown.
| Confirmed (Q3 2026 earnings materials) | Unconfirmed / Speculative |
|---|---|
| Early production started at Fremont | Exact number of units produced so far |
| Target: outside deliveries around end of 2027 | Consumer availability before end of 2027 |
| Target: 20,000 units/week, at least a year out | A firm date for reaching 20,000 units/week |
| Line sized for ~1 million units/year capacity | Whether that full capacity will be used by 2027 |
| AI5 chip planned for integration into Optimus | Exact AI5-to-Optimus integration date |
| Design, AI, reliability issues targeted for end of 2027 | A confirmed $20,000-$30,000 retail price (this is a prior Musk target, not a published price) |
| Initial use cases: industrial, dangerous, repetitive, 24/7 work | Whether first buyers will be external customers or controlled pilot partners |
Market and Investor Reaction
The Optimus update did not arrive in isolation. It was published alongside Tesla’s broader Q3 2026 results, which Yahoo Finance’s coverage noted included a 43% year-over-year drop in quarterly profit even as the Optimus production milestone was announced. That split, a profit decline paired with a long-horizon robotics bet, is precisely the kind of quarter that produces divided reactions: some investors and analysts treat Optimus as the company’s most important long-term value driver, a framing scanx.trade’s coverage of the call attributed directly to Musk’s own comments that robots would “drive most of value” for Tesla going forward, while others focus on the near-term financial picture and view Optimus commentary as a distraction from softer core results.
That tension is not new for Tesla, whose stock has long traded on a mix of current-quarter automotive and energy results alongside speculative, multi-year bets like Optimus, Full Self-Driving, and robotaxi services. What changed this quarter is that one of those speculative bets now has an actual year attached to it, 2027, which gives analysts a concrete point to measure progress against over the next several quarters rather than an open-ended “someday.”
What Could Derail the 2027 Timeline
Several factors could push the end-of-2027 target later, based on what Tesla itself has flagged. First, the company has explicitly tied outside deliveries to resolving design, AI, reliability, and operating-life issues simultaneously, a four-part dependency where any single unresolved item could delay the whole timeline. Second, the AI5 chip’s own production and integration schedule is an external dependency that sits partly outside Optimus’s own engineering process. Third, Tesla’s broader chip supply relationships, including the dynamics around AI5 and AI6 manufacturing, introduce supply-chain variables that are not fully within Tesla’s direct control.
Finally, there is the matter of Musk’s own hedged language. Phrases like “at least a year, maybe a little longer” and descriptions of the ramp as a “shallow slope” are not the language of a company confident in a fixed date; they are the language of a company trying to set expectations that leave room for delay without having to walk back a headline number later. That is a reasonable, even prudent, communication strategy, but it also means the end-of-2027 figure should be read as a target Tesla is working toward rather than a guarantee.
Five Predictions for Optimus Through 2027
- Tesla will likely report incremental weekly or monthly Optimus production figures on future earnings calls, using them as the primary proof points for progress toward the 20,000-units-a-week target, rather than announcing the final rate all at once.
- Expect Tesla to disclose AI5 chip integration progress for Optimus on a future earnings call before it discloses a firm consumer price, since compute capability appears to be a prerequisite for the reliability bar Musk has described.
- The first “outside” deliveries around the end of 2027 are more likely to go to a small number of industrial or controlled pilot partners than to an open consumer order book, consistent with the stated focus on dangerous, repetitive, and around-the-clock work.
- Competitive pressure from Figure AI, Boston Dynamics, Agility Robotics, 1X Technologies, and Unitree will intensify through 2027, but based on currently available reporting, none of them has committed to a comparable million-unit-scale annual production line, leaving Tesla’s manufacturing ambition distinct in scale even if its timeline slips.
- Expect continued scrutiny of Tesla’s broader financial results, including margin and profit trends, to shape how investors weigh each future Optimus update, meaning robot headlines and quarterly profit figures will likely keep arriving together rather than separately.
The Bigger Picture for Tesla’s Hardware Strategy
Optimus’s timeline update lands as Tesla is juggling several other major hardware pushes at once, from energy storage hardware that has been offsetting softer vehicle results to its broader manufacturing footprint. Musk’s framing of robots as poised to “drive most of value” for Tesla signals that the company increasingly wants to be understood as more than an automaker, a positioning that has implications well beyond this single earnings call, including how the company allocates capital, engineering talent, and factory floor space across competing priorities in 2027 and beyond.
For now, the concrete takeaway is narrower but still significant: Tesla has, for the first time, attached a specific year to Optimus’s transition from internal prototype to external product. Whether end-of-2027 holds will be one of the more closely watched corporate promises in hardware over the next five quarters.
Frequently Asked Questions
Is Tesla really going to sell Optimus robots next year?
Tesla expects to begin delivering Optimus to companies outside Tesla around the end of 2027, according to comments Elon Musk made on the Q3 2026 earnings call. Since that call took place in October 2026, 2027 is accurately described as “next year” from today’s vantage point, though deliveries are targeted toward the end of that year, not its start.
How much will the Tesla Optimus robot cost?
Tesla has not published a confirmed retail or enterprise price for Optimus as of its Q3 2026 earnings materials. A $20,000 to $30,000 figure circulating online reflects a prior target Musk has mentioned in past commentary, not a current published Tesla price.
Where is Tesla building Optimus?
Tesla has started early production of Optimus on a high-volume line at its Fremont, California factory, in manufacturing space that was previously used for vehicle production.
What is Tesla’s production target for Optimus?
Tesla’s stated target is 20,000 units a week, equivalent to roughly 1.04 million units a year. Musk said reaching that rate will take at least a year from the Q3 2026 call, and possibly longer, because the robot’s design is still evolving.
What chip will power Optimus?
Tesla has discussed plans to integrate its AI5 chip into Optimus, based on comments from the Q3 2026 earnings call transcript, with an approximate timing that is contingent on Tesla’s broader chip plans proceeding as expected.
Who will be Optimus’s first customers?
Tesla has not named any specific outside customer or order for Optimus. The company has said initial external use cases are expected to center on industrial, dangerous, repetitive, and around-the-clock work, but has not confirmed whether early units will go to arms-length customers or controlled pilot partners.
How does Optimus compare with other humanoid robots like Figure AI or Unitree?
Other companies, including Figure AI, Boston Dynamics, Agility Robotics, 1X Technologies, and Unitree, are also developing humanoid robots for industrial and assistance use cases. Based on currently available reporting, none has published a production-rate or delivery-date commitment matching the scale of Tesla’s roughly million-unit-per-year Fremont line.
Why did Tesla’s profit fall the same quarter it announced Optimus production?
Tesla’s Q3 2026 results, reported alongside the Optimus production milestone, showed quarterly profit down 43% year-over-year, according to Yahoo Finance’s coverage of the earnings report. The two developments were disclosed in the same shareholder update but reflect separate parts of Tesla’s business.
