SpaceX acquires xAI, Musk's trillion-dollar "internal cycle"

By: KuLi, Deep Tide TechFlow

On February 2nd, SpaceX announced the acquisition of xAI.

Post-merger valuation is $1.25 trillion, bringing Musk’s rockets, Starlink, X platform, Grok—all under one roof.

In a statement on his official website, Musk described the new company as “the most ambitious vertically integrated innovation engine on Earth and beyond.”

But publicly available data shows that xAI’s revenue last quarter was $107 million, with a net loss of $1.46 billion. In the first nine months, it burned nearly $10 billion in cash, averaging close to $1 billion per month. Meanwhile, SpaceX is projected to make about $8 billion in profit in 2025.

A company burning $1 billion a month, folded into a company earning $8 billion annually, and planning an IPO to raise $50 billion.

For xAI shareholders, this deal has a more practical name: a lifeline.

Left hand to right hand

This isn’t the first time Musk has shifted assets between his companies.

In March 2025, xAI acquired X. The merger was valued at $113 billion, which was widely considered an overestimate because X’s ad revenue has never recovered to Twitter-era levels. But the key point isn’t how much X is worth; it’s the real-time data generated by its 600 million active users, all fed into Grok’s training pipeline. This is xAI’s biggest exclusive resource compared to OpenAI and Anthropic.

In January 2026, Tesla invested $2 billion in xAI’s Series E funding round, citing the need to deeply leverage Grok’s capabilities for vehicle systems and Optimus robot training. Last year, Tesla shareholders voted on whether to invest in xAI; the majority approved, but the board ultimately did not. This time, the $2 billion investment was approved and executed.

On February 2nd, SpaceX acquired xAI. Because Tesla holds xAI shares, it indirectly owns a minority stake in SpaceX after the merger.

Meanwhile, xAI has spent hundreds of millions over the past year purchasing Tesla’s Megapack batteries to power the Colossus supercomputing center. Tesla also sources Starlink services from SpaceX to provide internet for vehicles.

X has become a “dowry” in Musk’s business empire—first marrying it to xAI for algorithm credibility, and now, with xAI, “marrying” it to SpaceX—just to complete the final narrative puzzle for the massive $1.25 trillion IPO in June.

Money and people circulate among Musk’s companies, a system known in investor circles as Muskonomy.

Tesla and xAI investor Ross Gerber spoke frankly: “It’s like a bunch of overvalued companies merging into a bigger overvalued chaos, run by Elon. But from another perspective, it’s now purely Musk’s concept stocks. Want to invest in Elon? Here it is.”

In the public company system, a single controlling person moving assets and contracts among affiliated companies usually triggers regulatory scrutiny. But Musk’s companies are almost all private. No public financial disclosures, no independent board oversight, and shareholders are a small group of VCs and sovereign funds, not as strict as public shareholders.

As long as the company isn’t listed, this game of left hand to right hand can continue.

But after the IPO, the story changes.

xAI’s life-and-death crisis

Why did SpaceX want to swallow xAI?

Behind the grand space narrative, the core reason is simple: to save itself.

$1 billion per month—that’s xAI’s current burn rate.

$33 million daily, $1.4 million hourly, $23,000 per minute. After reading this, xAI just burned another $60,000.

Where did the money go? Most of it was poured into Colossus—the supercomputing cluster built in Memphis, Tennessee. It currently has over 200,000 Nvidia H100 GPU equivalents, with a target power of 2 gigawatts. Just procuring chips and batteries has cost billions. Additionally, in the first nine months, nearly $160 million was spent on equity incentives, reflecting the fierce competition for AI talent.

But revenue growth can hardly match this level of spending.

xAI’s full-year 2025 revenue is projected at around $500 million, mainly from Grok API calls and X Premium subscriptions. Management’s guidance to investors is $2 billion in revenue in 2026 and profitability in 2027.

$2 billion sounds substantial, but OpenAI’s annualized revenue in 2025 already exceeded $20 billion. Even if xAI hits its targets, it’s only one-tenth of OpenAI’s size.

Founded less than three years ago, xAI’s valuation skyrocketed from zero to $250 billion, with at least six funding rounds.

The latest was in January this year, a Series E of $20 billion, with investors including Nvidia, Valor Equity Partners, and Qatar Investment Authority. Combined with the previous $5 billion debt financing arranged by Morgan Stanley, xAI has raised over $40 billion in total.

A company losing billions annually and earning only $500 million, with a valuation of $250 billion, has a price-to-sales ratio of 500.

An IPO alone would be hard for the secondary market to accept at this valuation.

A pattern in the 2025 US stock IPO market is that almost all companies’ trading prices after listing are below their last private funding valuation.

In contrast, SpaceX’s situation is entirely different.

It might be one of the most profitable private companies in the world. The only commercial rocket company routinely sending astronauts to the International Space Station in the US. Starlink’s revenue has already surpassed rocket launches, and it’s recurring revenue—900 million paying users, paying monthly. This is the most favored business model in the secondary market.

But SpaceX also faces challenges going public alone.

Wall Street’s valuation logic is brutal: rockets are valued based on cash flow, AI companies on imagination.

Referring to traditional aerospace giant Lockheed Martin, the market’s PE multiple has long been 20-30x; even with a “hard tech premium” of 50x for SpaceX, with its projected $8 billion profit in 2025, its market cap would hover around $400 billion. Even at a “new space economy” premium of 100x PE, it’s only about $800 billion.

But what about AI companies? OpenAI is seeking an $830 billion valuation despite losses; Anthropic is valued at $350 billion.

Elon Musk’s target is over $1 trillion.

Therefore, the financial logic of merging SpaceX and xAI is straightforward: xAI can’t sustain its valuation alone; SpaceX’s profits backstop it, creating a “profitable, growth-oriented, moat-protected” bundle.

For IPO underwriters, this is much easier to sell than separate listings.

But Musk also needs a narrative to connect the rocket company and AI company convincingly.

He found one: the space data center.

The space data center, the new story for IPO

In the merger announcement, Musk wrote: “AI progress depends on large ground data centers, which require enormous power and cooling. The global demand for AI’s electricity cannot be met by ground solutions alone, even in the short term, and would burden communities and the environment.”

This statement has a satirical background. xAI’s Colossus supercomputing center is built in Memphis, and local communities have been protesting its pollution. NAACP and environmental groups are preparing lawsuits.

Musk claims ground data centers burden communities, citing his own data center as an example.

His proposed solution is to move computing power into space.

Powered by solar energy, launched by SpaceX rockets, data transmitted via Starlink satellites.

Last Friday, SpaceX submitted an application to the FCC to authorize the launch of up to 1 million satellites to support the “orbital data center” plan.

“I estimate that within two to three years, the lowest-cost location for AI compute will be in space,” Musk said last month in Davos.

This vision is grand but still very early.

No company has yet operated a data center in space. All of xAI’s compute power is on the ground. Jeff Bezos’s Blue Origin has announced a similar space backbone network plan, and Google has a project called Suncatcher for space data centers. All are still in conceptual stages.

But an IPO doesn’t require a product to be operational; it needs a compelling enough story.

A standalone SpaceX IPO would focus on rockets and Starlink. That’s already very good, but growth potential is limited. The global commercial launch market is about that size annually, and Starlink’s user growth will eventually saturate. With xAI, the story becomes “AI + space infrastructure,” a trillion-dollar narrative.

Adding the vision of space data centers, the story becomes “human computing power’s future in orbit—we are the only company with rockets capable of sending it up.”

For underwriters and roadshow presentations, this layered storytelling makes a huge difference.

As for when space data centers will actually run, that’s a matter for after the IPO.

If feasible, xAI will have an infrastructure advantage that other AI companies cannot replicate. OpenAI rents cloud services from Amazon and shares profits with Microsoft; Google negotiates power supplies with states to meet environmental standards. Musk doesn’t need to; he has his own cloud in space.

From rocket launches (SpaceX), satellite networks (Starlink), AI training (xAI), content distribution (X platform), to applications (Tesla Autopilot, Optimus robot), the entire industry chain is under Musk’s control.

Tesla shareholders, the “Gas” on the road to Mars

In this capital game, there’s an invisible loser: Tesla shareholders.

Complaints from the Tesla shareholder community have reached a boiling point. Several investors on social media questioned: In 2020, Musk hinted that Tesla shareholders had priority subscription rights for SpaceX, but after xAI was established in 2023, Tesla’s AI team was poached by xAI, and compute resources were diverted. Now Musk is asking Tesla to invest $2 billion in xAI and to hold indirect stakes in SpaceX’s $1.5 trillion and xAI’s $250 billion valuations.

One investor calculated: In 2020, SpaceX was valued at $100 billion; now it’s $1 trillion—a tenfold increase. In 2023, xAI was valued at $10 billion; now it’s $250 billion—a 25-fold increase. The “priority subscription rights” have turned into “high-level bailout rights.”

Tesla itself is running out of money. It still has $44 billion in cash, but car sales have declined for two consecutive years. This week, Tesla announced a $2 billion investment in xAI and plans to double capital expenditure. Wall Street analysts predict that due to massive investments in AI infrastructure, Tesla may face a cash flow deficit of $5-7 billion in 2026.

The timeline is telling:

December 2025: xAI completes $20 billion funding, valuation $230 billion

January 2026: Tesla announces $2 billion investment in xAI

January 30, 2026: SpaceX applies to launch 1 million satellites

February 2, 2026: Announces acquisition of xAI

A series of rapid moves within 60 days. Musk is playing a big game; Tesla shareholders are just pawns.

In Musk’s “Muskonomy” system, resources flow between companies, each transfer creating new valuation peaks.

But Tesla shareholders realize: their technology is being drained, funds are being called upon, and they can only “buy back” these assets indirectly at valuations far above what was initially promised.

On the other hand, Tesla will benefit from Musk and SpaceX’s success—it’s a pure SpaceX concept stock.

A vivid analogy:

“Tesla shareholders are now like a former wife whose ex-husband took her savings to start a new business. They curse Musk for illegal fund diversion, but when they see the ex-husband (SpaceX) about to launch a $1.5 trillion super IPO, they can’t help but rush to remarry, hoping to get a seat at that Mars-bound ticket—a complex mix of Stockholm syndrome and greed.”

Musk’s methodology’s further evolution

Musk gave a one-sentence mission statement for the merged company: “Create a conscious sun to understand the universe, extending the light of awareness to the stars.”

This sentence encapsulates Musk’s underlying logic: using an unfalsifiable grand vision to turn current financial issues into a long-term narrative.

SpaceX’s early survival was also driven by this. When rockets failed three times and the company nearly went bankrupt, the “humankind must become a multi-planetary species” vision supported the valuation.

Now, this methodology has been upgraded.

Instead of telling a story with one company, it’s about tying all companies together into a bigger story.

Rockets provide launch capacity, Starlink handles transmission, xAI offers intelligence, 𝕏 manages data, Tesla supplies energy and robotics. Each piece has flaws, but combined into a “space AI civilization” vision, the flaws become “unfinished links.”

Musk discovered a secret: in the private company stage, valuation is mainly driven by narrative. As long as the story is big enough and far-reaching, investors are willing to believe. SpaceX’s valuation rose from billions to $800 billion; Tesla from near bankruptcy to $1.6 trillion—all based on this logic.

But individual company narratives have a ceiling. Rockets can only talk about Mars; EVs about autonomous driving. Once hitting that ceiling, valuation growth stalls.

The solution: connect all narratives into a super-narrative.

In this story:

SpaceX is “space infrastructure operator,” xAI is “pioneering human compute in space,” Tesla is “carrier of robotics and energy ecosystems,” 𝕏 is “training ground for real-time data.”

Each story alone has issues: SpaceX’s Mars colonization is distant, xAI’s Grok can’t beat ChatGPT or Claude, Tesla’s sales are declining, 𝕏’s ad revenue has plummeted.

But together, these issues become “a grand blueprint still under construction.”

Is this vision worth $1.25 trillion? The IPO pricing in mid-2026 will give an initial answer. Will AI boost the rocket’s valuation or drag down its IPO? Time will tell.

By then, a conscious sun will need to shine into the financial reports.