Tesla and SpaceX are, for the first time, letting outsiders look at the math behind a merger that Wall Street has treated as all but settled for months. On Tuesday, October 20, 2026, Elon Musk confirmed on a call tied to SpaceX’s quarterly business results that both companies have formed special committees of independent directors to study a possible combination. He also did something analysts have not been doing: he hedged. “Some people out there assume that a SpaceX and Tesla merger is a certainty. I think that would be overstating it,” Musk said, according to the fact pattern confirmed across multiple outlets covering the call.
The gap between how markets are pricing this deal and how its own CEO is describing it is the real story here, and it matters most in hardware. The merger case floated by Musk leans heavily on chip design, in-house silicon, and shared computing infrastructure, not just rockets and cars sharing a logo. That is the piece of this story that belongs in the hardware conversation, and it is the piece Wall Street’s “inevitable” framing tends to skip past.
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 and SpaceX Actually Confirmed on Tuesday
Start with what is actually on the record, because a lot of chatter around this story has outrun the facts. Tesla and SpaceX have formed special committees of independent directors to examine a potential combination, according to BBC News. Musk said the committees were formed at shareholders’ request, not his own initiative. Any transaction would still require shareholder approval at both companies, and Musk said the committees “have not reached agreement on terms of any transaction and there’s no guarantee that they will.”
That is a meaningfully different statement than “Tesla is buying SpaceX” or “SpaceX is buying Tesla.” Both framings have circulated in headlines and on social media over the past 48 hours, and neither is confirmed by anything Musk or either company has said. What is confirmed: two boards have opened a formal review process, Musk is not in the room for it, and no valuation, structure, or exchange ratio exists yet. Everything past that point is market speculation layered on top of a real but early corporate process, a distinction Tesla North’s own headline captured by framing the news as committees forming, not a deal closing.
| Fact | Status as of October 20, 2026 |
|---|---|
| Special committees of independent directors | Confirmed formed at both Tesla and SpaceX |
| Why the committees were formed | Musk says at shareholders’ request |
| Agreed deal terms | None reached; no guarantee of agreement |
| Musk’s role in committee talks | Recused, citing conflicted-party status |
| Tesla shareholder vote threshold | 50% of all outstanding shares, per Musk |
| Deal completion date | Unconfirmed; no timeline announced |
| SpaceX acquiring Tesla, or vice versa | Unconfirmed; structure not decided |
Musk’s Own Words: “Overstating It”
Musk has been unusually direct about pushing back on deal-certainty narratives, which is notable given how often he has talked up ambitious combinations in the past. Asked directly whether he thought merging the two companies made sense, he did not dodge the question. “Do I think it would be a good idea? Yes,” he said, according to BBC News. But in the same breath he drew a hard line between his personal opinion and what the companies will actually do, repeating that the decision sits with the independent committees and that no agreement exists.
He also addressed why he is not steering the process himself. “Because I’m a conflicted party, I can’t actually be in the discussion of the special committees,” Musk said, per reporting carried on Fidelity’s newswire feed. That is a standard governance move for a founder-CEO who sits atop both companies and stands to gain or lose from however the terms shake out, but it also means the person most publicly associated with “inevitability” has formally recused himself from deciding whether it happens.
Inside the 50% Vote Threshold
The detail that should matter most to anyone modeling deal odds is procedural, not emotional. Musk said approval on the Tesla side would require a vote representing 50% of all outstanding shares, not simply a majority of the shares that actually get voted. He called that “quite a high hurdle to achieve.” That distinction is easy to miss but changes the math substantially. A majority-of-votes-cast threshold can pass with enthusiastic turnout from a minority of the share base. A majority-of-all-shares threshold effectively requires near-full participation plus a real majority, which means abstentions and non-votes function as “no” votes rather than being excluded from the count.
For a company with Tesla’s shareholder base, which includes a large contingent of retail investors who do not reliably vote in corporate actions, clearing 50% of all shares outstanding is a genuinely different exercise than clearing 50% of votes cast. It is the kind of threshold that has sunk ambitious corporate combinations before, and it is also the threshold Musk chose to highlight himself rather than letting analysts assume a routine shareholder nod.
The Hardware Case: Chips, Not Just Capital
Set the corporate mechanics aside and the more interesting question for a hardware audience is what Musk says the combination is actually for. He has been specific that the rationale is not primarily about combining a car company’s balance sheet with a rocket company’s balance sheet. It is about chip design, computing infrastructure, robotics, and connectivity sitting under one roof instead of two separate governance structures negotiating supply deals with each other at arm’s length.
That framing lines up with a project that has already been publicly named: Terafab, described in multiple reports as a planned chip-manufacturing effort jointly tied to Tesla and SpaceX. Readers who followed our earlier coverage of the Terafab chip plant’s Q3 2028 production target already know the broad outline. What Tuesday’s committee news adds is the corporate logic for why that plant might eventually sit inside a single combined entity rather than being a cross-licensing arrangement between two separately owned companies.
Terafab: The Chip Plant Behind the Chip Talk
Terafab is worth pulling apart because it is the most concrete piece of hardware evidence behind an otherwise abstract merger pitch. According to reporting from outlets including TechCrunch and CNBC covering Tuesday’s SpaceX earnings call, Musk described a two-stage timeline: a first research-scale chip out of the facility as early as mid-2027, with substantial schedule uncertainty attached to that date, followed by a push toward volume production by the third quarter of 2028. Musk reportedly described the facility’s long-term ambition in characteristically large terms, framing its eventual output around what he called roughly a terawatt of “superintelligence chips” per year.
None of that is manufacturing capacity that exists today. It is a roadmap, and roadmaps for advanced chip fabs have a long history of slipping, including at far better-capitalized incumbents. Reports also describe Tesla, SpaceX, and Intel as collaborating on the chip work tied to Terafab, which, if accurate, would put a third major industrial partner inside the hardware thesis that supposedly justifies combining Tesla and SpaceX. That raises its own question: if Intel is already a partner on the silicon, how much of the “merge the companies” logic is really about manufacturing versus about simplifying who owns the intellectual property and the balance sheet risk of building a leading-edge fab from scratch.
AI5, AI6, and Dojo: Tesla’s In-House Silicon Problem
Tesla’s own chip ambitions predate this merger conversation by years. Dojo was the company’s attempt to build in-house AI-training infrastructure rather than depending entirely on outside accelerator vendors, and the AI5 and AI6 designations refer to Tesla’s newer custom silicon efforts aimed at both vehicle autonomy and, increasingly, broader AI workloads. It is worth being precise about what is and is not confirmed here: Tuesday’s merger coverage did not include new production volumes, process-node details, or launch dates for AI5 or AI6. Treat any specific AI5/AI6 spec you see elsewhere as background context rather than something confirmed in this week’s reporting.
What is clear is the strategic problem Tesla is trying to solve. Building competitive AI silicon in-house is capital-intensive, and Tesla has historically split that cost and that engineering burden across its automotive balance sheet alone. A combined entity with SpaceX’s cash generation and satellite-and-launch infrastructure behind it would, in theory, change that math. Whether that is worth giving up the governance simplicity of two separately traded, separately accountable public companies is exactly the question the special committees were created to answer, and exactly the question Musk has declined to pre-judge in public.
SpaceX’s New Role as an AI Compute Landlord
The other half of the hardware argument is less about chips SpaceX makes and more about computing capacity SpaceX now rents out. Reporting tied to Tuesday’s call describes SpaceX’s first reported quarterly profit as being driven in part by deals to lease computing power to AI companies, a detail that reframes SpaceX from a pure launch-and-satellite business into something closer to an AI infrastructure operator with rockets as a side business. That shift matters for the merger thesis because it means SpaceX is no longer just a potential customer for Tesla’s chip ambitions; it is becoming a data-center landlord in its own right, which changes what “combining the companies” would actually combine.
Our earlier coverage of SpaceX’s first profitable quarter and the company’s forward revenue guidance tied to AI compute both point in the same direction: SpaceX’s finances are increasingly intertwined with AI infrastructure demand rather than purely with satellite broadband subscriptions or government launch contracts. That is a genuinely new dynamic compared to how SpaceX was valued even two years ago, and it is a big part of why some analysts now treat combining it with Tesla as a logical next step rather than a vanity project.
Starmind and the Orbital Data Center Pitch
Musk’s own list of merger benefits extended past terrestrial chips into orbit. He pointed to what multiple reports describe as the Starmind constellation, a planned network of orbital data centers, as another area where combining Tesla’s compute and chip ambitions with SpaceX’s launch cadence and satellite platform could pay off. The pitch is straightforward on paper: if you are already launching large satellite constellations at a cadence no competitor can match, adding AI compute payloads to that same supply chain is a smaller incremental step than it would be for a company building a space program from zero.
It is also, again, a roadmap rather than a running system. Orbital data centers face thermal, radiation, and latency constraints that terrestrial facilities do not, and none of the reporting around Tuesday’s call quantified timelines, payload counts, or power budgets for Starmind specifically. The honest way to characterize it: this is a strategic direction Musk has publicly committed to pursuing, not a deployed product line that changes the merger’s near-term hardware math.
Competitive Comparison: How Other Tech Giants Vertically Integrate AI Hardware
The Tesla-SpaceX pitch is not happening in a vacuum. Every major AI infrastructure player has been moving toward some version of owning more of its own silicon stack rather than depending entirely on merchant chip vendors. The difference is that most of those efforts stay inside a single existing company rather than requiring two independently traded public companies to merge first. The table below lays out how the proposed Tesla-SpaceX combination compares to the vertical integration strategies already underway elsewhere in the industry.
| Company | In-house chip effort | External compute reliance | Vertical integration angle |
|---|---|---|---|
| Tesla-SpaceX (proposed) | Dojo, AI5/AI6, Terafab (reported, with Intel) | Still dependent on merchant GPUs today | Vehicles, rockets, satellites, robotics under one roof |
| Alphabet/Google | TPU custom accelerators | Low, mostly self-supplied | Search, cloud, and Waymo autonomy on one chip line |
| Amazon | Trainium and Inferentia | Moderate, plus Anthropic partnership | AWS cloud plus AI model investment |
| Microsoft | Maia accelerators | High, anchored by OpenAI partnership | Azure cloud plus OpenAI model access |
| Meta | MTIA accelerators | Moderate, mixed with merchant GPUs | Social platforms and AR/VR hardware |
The pattern that stands out: Tesla and SpaceX are trying to reach the same destination as Alphabet or Amazon, owning more of the chip stack that feeds their AI ambitions, but doing it by merging two already-massive, already-public companies rather than building the capability inside one. That is a harder path structurally, even if the underlying engineering goal is identical to what every other hyperscaler is already chasing.
Historical Context: Musk’s Last “Inevitable” Merger
Tesla shareholders have been here before, and the parallel is instructive. In 2016, Tesla acquired SolarCity, a deal Musk also framed as strategically obvious and that also drew significant shareholder skepticism before it closed. That transaction later became the subject of years of litigation over whether the board adequately weighed the conflicts involved in a Musk-led company acquiring another Musk-led company. The SolarCity deal did ultimately close, but only after a contentious vote and years of subsequent legal scrutiny of how independent the board’s review really was.
The Tesla-SpaceX situation is both similar and different. It is similar in that Musk again sits on both sides of a potential transaction and has again recused himself from the formal committee process while still commenting publicly on the merits. It is different in scale: SolarCity was a struggling solar installer being absorbed into a much larger automaker, while a Tesla-SpaceX combination would join two of the most valuable private-and-public companies in the world, with SpaceX’s valuation built substantially on private funding rounds rather than public market pricing. That makes the valuation and exchange-ratio problem considerably harder to solve than anything the SolarCity deal faced.
Market Reaction and Analyst Positioning
Tesla stock has moved on merger headlines before, and the pattern around this latest round of committee news has been no exception, building on the reaction we tracked when merger talk first drove an 8% single-session jump in Tesla shares. The challenge for anyone trying to price this deal is that almost none of the inputs that normally drive a merger valuation are public yet: no exchange ratio, no agreed structure, and crucially, no independent valuation of SpaceX itself, which does not trade on public markets and has historically been valued through private secondary transactions that move in large steps rather than continuously.
That has not stopped analysts from treating the combination as directionally likely, consistent with the broader narrative we covered when merger-bid chatter put a combined entity in rarefied valuation territory. The gap between that analyst framing and Musk’s own “overstating it” comment on Tuesday is exactly the tension this story turns on: markets are pricing in a high probability of eventual combination, while the person with the most information about the actual committee discussions is the one publicly cautioning against treating it as decided.
The Chip Supply Chain Still Needs to Prove Itself
Strip away the corporate drama and the hardware case for this merger rests on a chip supply chain that does not exist in volume yet. Terafab’s research-chip target is still roughly seven months out from today, and its volume-production target is nearly two years away. Our earlier look at the broader AI chip substrate buildout underway across the industry shows how capital-intensive and timeline-sensitive this category is even for established suppliers, let alone a brand-new fab effort tied to a not-yet-approved corporate merger. Nothing in Tuesday’s reporting shows SpaceX is already manufacturing Tesla’s AI5 or AI6 chips at scale, and nothing shows a completed, shared GPU or chip cluster spanning both companies today.
That is the detail that should temper the “inevitable” framing most. Wall Street’s certainty is partly a bet on a merger happening and partly a bet on a chip-manufacturing roadmap executing on schedule, and those are two separate things that both have to go right. Musk’s own public caution reads, in that light, less like false modesty and more like an accurate description of how much is still unbuilt.
What Musk and Analysts Are Saying
Musk’s own comments on Tuesday are the clearest window into how he is framing the hardware rationale versus the deal uncertainty. Asked for his personal view on the idea, he was unambiguous: “Do I think it would be a good idea? Yes,” Musk said, according to BBC News.
But he immediately qualified that personal view by pointing back to the committee process: “The special committees have not reached agreement on terms of any transaction and there’s no guarantee that they will,” Musk said, per the same BBC report.
On why he is not driving that process himself, Musk was equally direct: “Because I’m a conflicted party, I can’t actually be in the discussion of the special committees,” he said, according to Bloomberg Law’s coverage of the committee formation.
And on the hardware rationale specifically, Musk laid out the fullest version of the pitch: “I do think there’s a lot of merit to combining the companies as we are pretty much in the superintelligence singularity, and there’s a lot to be gained from cooperating on developing superintelligence software, chip design, the Terafab obviously, robotics, mass-to-orbit with the Starmind constellation (of orbital data centers), connectivity,” Musk said, per reporting carried on Fidelity’s newswire feed.
Risk Factors That Could Still Sink the Deal
Beyond the headline 50%-of-all-shares threshold, several practical risk factors sit underneath this process. SpaceX does not trade publicly, so any exchange ratio requires agreeing on a valuation for a company whose last marks came from private secondary sales rather than continuous market pricing, a notoriously contentious exercise even in friendlier deals. Tesla’s board also faces the same conflicted-founder scrutiny that dogged the SolarCity acquisition for years afterward, meaning any agreed terms will likely draw immediate legal challenge regardless of how the vote goes. And the hardware roadmap itself, Terafab’s chip timeline, is exactly the kind of multi-year industrial buildout that has a well-documented history of slipping at companies with far more fab experience than either Tesla or SpaceX currently has.
Put together, those factors describe a process that could easily run past the committees reaching any agreement at all, let alone past a shareholder vote clearing a 50%-of-all-shares bar. None of that makes a deal impossible. It does make “inevitable” a stronger word than the facts currently support.
Predictions: Where This Likely Goes Next
- The special committees will take months, not weeks, to produce agreed terms, given that no valuation framework for the privately held SpaceX exists yet.
- Expect continued volatility in Tesla shares tied to merger headlines, with swings in both directions as committee updates, leaks, or denials surface.
- Terafab’s mid-2027 research-chip target is likely to slip at least somewhat, consistent with how most first-generation advanced chip fab timelines move in this industry.
- Regardless of whether the full merger closes, expect Tesla and SpaceX to deepen commercial chip and compute agreements between the two companies in the interim, since that cooperation does not require shareholder approval.
- Any final terms that do emerge will likely draw shareholder litigation almost immediately, echoing the years of legal scrutiny that followed the 2016 SolarCity acquisition.
Frequently Asked Questions
Has a Tesla-SpaceX merger actually been agreed to?
No. As of October 20, 2026, both companies have only formed special committees of independent directors to examine the idea. Musk has said no terms have been agreed and there is no guarantee a transaction will happen.
Why does Musk say a merger is not certain?
Musk has said he personally favors the idea but that the decision rests entirely with the independent special committees, which have not reached agreement on terms. He has specifically pushed back on the idea that a deal should be treated as a certainty.
What vote threshold would Tesla shareholders need to clear?
Musk said approval on the Tesla side would require a vote representing 50% of all outstanding shares, not just a majority of the shares actually voted, which he described as a high hurdle.
Is Musk involved in the special committee discussions?
No. Musk has said he is a conflicted party given his leadership roles at both companies and is not participating in the committee discussions.
What is Terafab and how does it relate to the merger?
Terafab is a planned chip-manufacturing project reported to be jointly tied to Tesla and SpaceX, reportedly developed with Intel involvement, with a research-chip target around mid-2027 and a volume-production target around the third quarter of 2028. It is one of the concrete hardware projects cited as a rationale for combining the companies.
Does this merger involve Tesla’s AI5 and AI6 chips directly?
Tuesday’s reporting did not confirm specific production volumes, specs, or timelines for AI5 or AI6 tied to the merger discussion. They remain part of Tesla’s broader in-house chip strategy but were not detailed in this week’s committee news.
How is this similar to Tesla’s 2016 SolarCity acquisition?
Both deals involve Musk sitting on both sides of a potential transaction and both drew early shareholder skepticism. The SolarCity deal closed but faced years of subsequent litigation over board independence, a precedent likely to shape scrutiny of any Tesla-SpaceX terms that eventually emerge.
When will a final decision be made?
No timeline has been confirmed. Given that no valuation framework exists yet for the privately held SpaceX, and that complex multi-billion-dollar valuations typically take months to negotiate, a near-term resolution looks unlikely.
