Goldman Sachs Expects SpaceX AI Revenue to Grow 100x to $322 Billion by 2030
Alina Collins
Goldman Sachs's model, presented to potential investors as lead IPO underwriter, projects SpaceX's AI division revenue surging from $3.2 billion in 2025 to $322 billion by 2030 — a roughly 100-fold increase that reframes the rocket company's future around AI, not space.
Where does the $322 billion AI revenue assumption come from?
The model's foundation: SpaceX's AI arm xAI is assigned a total addressable market (TAM — the theoretical ceiling for the business) of $26.5 trillion.
By comparison, Starlink satellite internet plus rocket launch services carry a combined TAM of roughly $2 trillion — the AI market assumption is more than 13 times larger.
This means → Goldman's valuation story for SpaceX has a new protagonist: not satellites, not rockets, but AI.
How steep is the projected growth path?
Goldman's AI revenue trajectory: $3.2 billion in 2025 → $15.6 billion in 2026 (up 388% year-on-year) → $34.5 billion in 2027.
By 2030, total SpaceX revenue is projected at $474 billion, with AI contributing close to two-thirds.
In plain terms = the curve Goldman has drawn requires SpaceX's AI business to grow from a fledgling division into the company's dominant revenue engine within five years — doubling every year would not be enough.
Will the existing businesses be sidelined?
Starlink satellite internet is projected at $144 billion in 2030 revenue — less than half of the AI division.
Rocket services are projected at $8.3 billion in 2030, up from $4.1 billion last year — a modest double.
This reflects a stark judgment inside the Goldman model: SpaceX's core narrative is shifting from "space-infrastructure company" to "AI company."
How aggressive are the profitability forecasts?
Goldman projects SpaceX's adjusted EBITDA — earnings before interest, taxes, depreciation, and amortization, a measure of cash-generating power — rising from $6.6 billion in 2025 to $352 billion by 2030.
The company posted negative $13.8 billion in free cash flow last year; Goldman expects it to turn positive only in 2031, reaching $72 billion.
This means → for the next five to six years, investors are not buying profit — they are buying a belief in explosive AI growth.
Can this story hold up?
The Financial Times notes that sustaining these projections requires Musk's Grok model family to catch up with and surpass Anthropic, Google, and OpenAI in programming, cybersecurity, AI agents, and chatbots.
The IPO roadshow is underway; the deal could raise up to $86 billion. Goldman won the lead-underwriter mandate over Morgan Stanley, JPMorgan, Citi, Bank of America, and UBS.
In plain terms = Goldman is both the architect of the valuation model and one of the biggest beneficiaries of this deal — the referee and the player are the same person, and investors will need to judge the forecast's credibility for themselves.
Content is for reference only, not financial advice.