Tesla told investors on October 21, 2026, that it is pulling back on capital spending for the rest of the year, even as CEO Elon Musk doubles down on artificial intelligence, humanoid robots, and a possible merger with SpaceX. The company cut its 2026 capital-expenditure forecast to roughly $20 billion, down from a prior guidance of more than $25 billion, according to Reuters. The move lands just as Tesla and SpaceX have formed special committees to study a combination of the two companies, a deal Musk called “a good idea” without promising it will happen.
The timing is the story here. Tesla is cutting cash outflows at the exact moment its future, by Musk’s own description, depends on spending tens of billions more on AI compute, Optimus robots, and a custom chip pipeline. That tension between near-term capital discipline and long-term AI ambition is now central to how investors, analysts, and rivals are reading Tesla’s third-quarter 2026 results.
Don't miss new tech stories on Google
Add Tech Insider once in the Google app and our stories appear in your news suggestions.
Tesla’s Capex Guidance Cut, by the Numbers
Tesla’s third-quarter capital expenditure came in at $4.7 billion for the quarter ended September 30, 2026. That figure was more than double what Tesla spent a year earlier, but it fell well short of the $7.5 billion that analysts surveyed by LSEG had expected, according to the BBC. On the back of that quarterly undershoot, Tesla revised its full-year 2026 capex guidance down to approximately $20 billion, from a previous outlook of more than $25 billion.
Tesla did not detail which specific programs account for the reduction. The company has previously said future capital spending would expand to cover robotaxi fleet build-out, Optimus production capacity, solar manufacturing, AI-compute infrastructure, and a semiconductor fab, but it stopped short of saying which of those categories moved slower than planned in the third quarter. That ambiguity is itself notable: Tesla is simultaneously telling investors it spent less than expected while describing an AI roadmap that assumes it will need to spend far more in the years ahead.
| Metric | Prior Guidance / Estimate | Reported Q3 2026 Figure |
|---|---|---|
| Full-year 2026 capex guidance | More than $20 billion* (revised) | ~$20 billion (new guidance) |
| Full-year 2026 capex guidance (original) | More than $25 billion | Revised down to ~$20 billion |
| Q3 2026 capital expenditure | $7.5 billion (LSEG analyst average) | $4.7 billion (actual) |
| Q3 2026 free cash flow | Not separately forecast | Positive, more than $1.5 billion |
| Cash and investments at quarter-end | Not separately forecast | More than $45 billion |
Tesla ended the quarter with more than $45 billion in cash and investments, giving the company a sizable cushion even as capital spending slows. Reuters reported the positive free cash flow came in part from lower-than-planned AI-related spending rather than from a jump in underlying profitability, a distinction that matters for anyone trying to read Tesla’s quarter as a sign of operational strength versus simple capital-timing.
The Tesla-SpaceX Merger Talks, Explained
The capex news arrived alongside the bigger headline from Tesla’s earnings call: Musk confirmed that Tesla and SpaceX have each formed special committees of independent directors to examine a potential merger between the two companies. Musk told analysts he had “no further comment” on the merger during the call itself, but his broader remarks made clear the idea is alive. “I do think there’s a lot of merit to combining the companies,” Musk said, according to a transcript of the earnings call reported by Fidelity.
Musk was careful to frame the process as exploratory rather than decided. He said the special committees had not reached agreement on transaction terms and that there was “no guarantee” a deal would proceed. He also said there was “a lot of merit to merging and a lot to be gained,” language that signals personal enthusiasm for the idea even as the formal corporate process remains at an early stage.
What makes this merger talk different from ordinary speculation is the existence of the special committees themselves. Public companies typically form independent committees specifically when a transaction involves a conflict of interest, and Musk is CEO of both companies. That structure suggests the conversation has moved beyond hypothetical musing into something resembling a formal evaluation process, even though no terms, valuation, or timeline have been disclosed.
Why the 50% Shareholder Vote Is a Real Obstacle
Musk used his clearest language of the call to flag the regulatory and governance hurdle any Tesla-SpaceX combination would face. He said a merger would require approval from at least 50% of all outstanding Tesla shares, not just 50% of shares actually cast in a vote. He called that threshold “quite a high hurdle to achieve.”
That distinction is significant for anyone trying to handicap the odds of a deal closing. A simple majority of votes cast is a far lower bar than a majority of every share outstanding, because many retail shareholders never vote at all. Tesla has a large base of individual investors, and getting more than half of all outstanding shares to actively approve a related-party transaction, where the CEO sits on both sides of the table, is a meaningfully harder governance test than a typical merger vote.
Tesla shareholders are already scheduled to vote on a separate, high-profile governance matter: a December 3, 2026 meeting to decide whether Musk remains on Tesla’s board, a vote detailed in Tesla’s December 3 shareholder vote announcement. Any merger vote would likely follow a similarly contested path, given how closely Tesla’s retail shareholder base tracks Musk-related governance questions.
Musk’s AI and Robot Bet Keeps Growing
The capex cut does not signal retreat from AI. If anything, Tesla’s own disclosures point the opposite direction. Research spending rose by $1.13 billion during the third quarter as the company pushed further into AI development, a figure larger than Tesla’s $718 million operating profit for the period. Musk said Tesla was making “good progress” on the AI systems controlling Optimus humanoid robots, and the company has outlined plans, subject to execution, to integrate its AI5 chip into Optimus around April 2027.
Musk has previously described the scale of compute spending he believes humanoid robotics will eventually require. “Like, ultimately, it will be tens of billions of dollars on training compute,” Musk said at Tesla’s 2025 shareholder meeting, a remark preserved in a transcript published by WhatsUpTesla. He also said he has personally spent significant time on Tesla’s custom AI chip design, noting that “in order to have a functional robot, you have to have a great AI chip.”
Tesla has said it expects its driverless robotaxi fleet to begin contributing financially in 2027. That timeline sits alongside Tesla’s broader robotaxi push, which has included expanding service areas and working through regulatory scrutiny, including a federal probe covered in NHTSA’s review of roughly 1,000 Cybercab units. None of that work is cheap, which is why the near-term capex reduction reads less like austerity and more like sequencing: Tesla appears to be holding off on some spending in 2026 while the AI5-Optimus-robotaxi pipeline matures, rather than cutting the long-term AI roadmap itself.
Free Cash Flow: A More Complicated Story Than It Looks
Tesla’s positive free cash flow for the quarter, more than $1.5 billion, is the headline number bulls are likely to point to. But the composition of that figure matters. Reuters’ reporting attributes the positive swing largely to lower-than-planned AI-related capital deployment rather than to a surge in underlying profitability. Tesla’s operating profit of $718 million was smaller than the $1.13 billion increase in research spending alone, underscoring how much of Tesla’s current earnings picture is being shaped by AI investment decisions rather than core vehicle economics.
That nuance matters for the SpaceX merger conversation too. If Tesla’s free cash flow strength is partly a function of delayed AI capex rather than durable operating improvement, it becomes a weaker data point for valuing Tesla’s side of any future merger math. Investors evaluating a potential combined entity will need to separate cash flow that reflects real margin improvement from cash flow that simply reflects spending Tesla has pushed into future quarters.
How Wall Street and the Market Reacted
News of the special committees moved markets for both companies. SpaceX shares reportedly rose in after-hours trading following Musk’s confirmation that the merger process had formally begun, a reaction consistent with the market’s prior response to merger developments detailed in earlier coverage of an 8% stock jump tied to merger talks. Tesla’s revenue and deliveries for the quarter beat estimates, and the combination of a capex cut, positive free cash flow, and merger confirmation gave investors multiple angles to trade on at once.
Analysts had already been weighing how AI spending ambitions at both companies would intersect with merger speculation, a dynamic flagged ahead of the earnings report in coverage of how the two companies’ results would test Musk’s AI spending ambitions as merger talk grew. The capex guidance cut adds a new variable: it suggests Tesla’s board and management are being more conservative about 2026 cash outflows than the market expected, even as the long-term AI story stays aggressive.
Competitive Comparison: Tesla’s AI Capital Strategy Against Big Tech
Tesla is not alone in facing investor scrutiny over AI capital spending. Meta Platforms raised its own 2026 AI capex outlook to roughly $145 billion even as Tesla was cutting its guidance in the opposite direction, a contrast detailed in coverage of Meta’s capex hike against Tesla’s cuts. That divergence highlights how differently large technology companies are managing the AI investment cycle heading into 2027: some are accelerating spend to secure compute capacity, while Tesla is trying to demonstrate capital discipline while still claiming an aggressive AI roadmap.
| Company | 2026 Capex Direction | Primary AI Spending Focus |
|---|---|---|
| Tesla | Cut from >$25B to ~$20B guidance | Optimus robots, AI5 chip, robotaxi fleet, AI compute |
| Meta Platforms | Raised to ~$145B guidance | AI data centers, model training infrastructure |
| SpaceX | Growing, reportedly profitable | Starlink, AI compute services, launch cadence |
The comparison matters for the merger discussion because it frames what kind of partner SpaceX would be bringing to the table. SpaceX has been expanding its own AI-adjacent revenue streams, a trend covered in reporting on SpaceX’s AI revenue climbing to $7.56 billion. If Tesla is pulling back on capex while SpaceX’s AI-linked business keeps expanding, that asymmetry could factor into how any merger is eventually valued and structured.
Historical Context: From Expansion Mode to Capital Discipline
Tesla has spent much of the past two years in heavy expansion mode, building out Gigafactories, energy storage lines, and AI compute clusters while repeatedly telling investors that near-term spending would pay off in future robotaxi and robotics revenue. The $25 billion-plus original 2026 capex guidance fit that pattern. The revision down to roughly $20 billion marks one of the more visible moments in 2026 where Tesla has pulled back on a prior spending commitment rather than expanded it.
That shift follows a third quarter in which Tesla’s broader financial picture showed mixed signals: strong energy storage growth and beaten revenue estimates alongside a decline in automotive profitability, a pattern documented in Tesla’s Q3 2026 earnings breakdown covering 486,532 vehicle deliveries and 13.7 GWh of storage deployments. Tesla’s Optimus production ramp has also faced its own financial strain, reflected in a separate report on Tesla starting Optimus production as profit fell, covered in Tesla’s Optimus production launch amid a 43% profit decline. Seen together, the capex cut looks less like an isolated decision and more like part of a broader effort to manage cash through a period where AI and robotics investment is weighing heavily on reported profit.
What a Merger Would Actually Mean for Shareholders
A Tesla-SpaceX combination, if it happens, would create one of the most unusual corporate structures in recent memory: a publicly traded automaker and energy company merging with a privately held rocket and satellite-internet business. Tesla shareholders would be voting not just on a standard merger but on bringing SpaceX’s capital needs, launch-business risk profile, and Starlink growth story directly onto Tesla’s public balance sheet.
The valuation and structure questions remain entirely open. No terms have been disclosed, and Musk has been explicit that the special committees have not agreed on anything. Given earlier merger-adjacent reporting discussed Tesla’s position in a potential combined entity’s valuation ranking, referenced in coverage of a merger bid targeting a $3.5 trillion combined valuation, any deal would likely be scrutinized heavily by both regulators and independent shareholders precisely because of the conflict-of-interest dynamics built into having Musk run both sides.
Analyst and Industry Reaction
Tesla described its Optimus manufacturing milestone in stark terms during the earnings materials, noting that “the robots coming off this line are the first volume-production robots we have manufactured,” according to a report from the Seoul Economic Daily. That framing is central to how Tesla wants investors to read this quarter: not as a company slowing down, but as one transitioning from heavy build-out spending toward early production revenue.
Musk also pushed back on the idea that a merger conversation was anything beyond preliminary, telling analysts at one point that “we can’t talk about combining companies and that kind of thing” in reference to the formal limits the special-committee process places on what executives can say publicly before terms are set. That caution is likely to frustrate investors looking for clarity, but it is also standard practice for any transaction involving related parties and independent director review.
Risks and Open Questions
Several questions remain unresolved heading into the next few months. Tesla has not specified which programs absorbed the capex reduction, leaving open whether robotaxi expansion, Optimus scaling, AI data-center buildout, or the semiconductor fab plan slowed down. The special committees have not published a timeline for reaching a merger recommendation, and Musk has given no indication of when shareholders might see concrete terms to vote on. The 50%-of-all-outstanding-shares threshold also means a merger could clear the special committee process internally and still fail at the shareholder vote if turnout and support fall short.
There is also the question of whether Tesla’s capital discipline this quarter is a durable shift or a temporary pause. Tesla’s own long-term messaging, including Musk’s prior comments about tens of billions of dollars eventually going toward AI training compute, suggests spending will rise again once Optimus and robotaxi programs move from early production into scale. Investors will be watching Tesla’s fourth-quarter 2026 results closely to see whether the $20 billion full-year figure holds or gets revised again.
Five Predictions for What Happens Next
- Tesla’s 2027 capex guidance will likely come in higher than 2026’s revised $20 billion figure, as AI5 chip integration into Optimus and robotaxi fleet expansion both move toward scale production.
- The Tesla-SpaceX special committees will take multiple quarters, not weeks, to reach a recommendation, given the complexity of valuing a private aerospace company against a public automaker.
- Any formal merger proposal will face vocal opposition from a segment of Tesla’s retail shareholder base, making the 50%-of-outstanding-shares threshold a genuine risk to deal completion rather than a formality.
- Tesla will continue to frame near-term spending restraint as a bridge to robotaxi and Optimus revenue starting in 2027, rather than as a retreat from its AI ambitions.
- Expect continued scrutiny from regulators and analysts over how much of Tesla’s reported free cash flow reflects delayed AI capex versus genuine operating improvement, especially if the merger process advances.
Frequently Asked Questions
Why did Tesla cut its 2026 capital expenditure forecast?
Tesla revised its 2026 capex guidance down to approximately $20 billion from a prior forecast of more than $25 billion after third-quarter capital spending came in at $4.7 billion, below the $7.5 billion analysts surveyed by LSEG had expected. Tesla has not detailed exactly which programs account for the lower spending.
Is Tesla merging with SpaceX?
No merger has been finalized. Tesla and SpaceX have formed special committees of independent directors to examine a potential combination, but Musk has said the committees have not agreed on terms and there is no guarantee a deal will happen.
What shareholder approval would a Tesla-SpaceX merger need?
Musk said any merger would require approval from at least 50% of all outstanding Tesla shares, not simply 50% of the shares that are actually voted. He described that threshold as a high hurdle to clear.
Does the capex cut mean Tesla is reducing its AI investment?
Not based on current disclosures. Tesla’s research spending actually rose by $1.13 billion during the third quarter, and the company continues to describe plans for AI5 chip integration into Optimus robots and expansion of its robotaxi fleet. The capex guidance reduction reflects lower overall capital deployment, not a stated pullback from AI programs specifically.
How much free cash flow did Tesla generate in the third quarter of 2026?
Tesla reported positive free cash flow of more than $1.5 billion for the quarter, with cash and investments totaling more than $45 billion at quarter-end.
When does Tesla expect its robotaxi fleet to become profitable?
According to Musk’s earnings-call comments, Tesla expects its driverless robotaxi fleet to begin contributing financially in 2027.
What is the AI5 chip and why does it matter for Optimus?
The AI5 chip is Tesla’s custom-designed processor intended to power its AI systems. Tesla has outlined plans, subject to execution, to integrate the AI5 chip into Optimus humanoid robots around April 2027, a step the company considers central to making the robots functional at scale.
How did markets react to the capex cut and merger news?
SpaceX shares reportedly rose in after-hours trading following confirmation of the special committees, while Tesla’s own results, which beat revenue estimates despite a decline in automotive profitability, gave investors multiple data points to weigh at once.
