Goldman Sachs Just Made a Bold Statement About SpaceX. Here's Why I'm Not Buying It

Goldman Sachs, the lead underwriter for SpaceX's planned June 12 IPO, projects SpaceX's AI division could reach $322 billion in revenue by 2030, up from $3.2 billion in 2025. It also forecasts total SpaceX revenue of $474 billion by 2030 versus $18.7 billion last year, supporting a $1.77 trillion valuation at $135 per share for the offering. The piece is largely a commentary on aggressive IPO marketing assumptions rather than a confirmed operating update.

Analysis

The market is being asked to underwrite not just a new listing, but a narrative reset: the underwriting bank has a strong incentive to anchor valuation expectations with a very aggressive long-dated growth story. That matters because when a transaction is priced at a multiple that implies perfection, the first secondary effect is not on the issuer itself but on every adjacent “picks and shovels” name used to justify the story—AI compute, networking, and semiconductor infrastructure can all get bid as investors extrapolate a larger addressable market.

The more interesting read-through is that the hype itself may be a better near-term trade than the fundamentals. If institutions accept a 2030 ramp that implies extreme revenue acceleration, they are effectively pricing optionality on AI monetization rather than current cash generation, which tends to lift the entire private-market complex and late-stage crossover names. That can be bullish for AI leaders in the tape, but it also raises the odds of a post-deal air pocket if the first lockup period reveals that the market is paying for story, not visible backlog.

The contrarian risk is that this becomes a sentiment top for mega-cap AI. When the market starts validating increasingly large revenue forecasts for a single headline company, capital can rotate from the proven incumbents into the speculative “next layer” of beneficiaries, compressing relative returns in names like MSFT and NVDA even if they remain structurally strong. In other words, the trade may be less “buy the obvious winners” and more “fade the second-order enthusiasm once the IPO clears.”

For GS, the short-term benefit is fee generation and positioning as the market-maker for a landmark transaction, but the medium-term risk is reputational if the roadshow assumptions look stretched. That creates a setup where the stock can outperform into pricing, then mean-revert if investors conclude the bank is stretching estimates to support valuation rather than delivering a balanced underwriting case.

Key Decisions for Investors

• Tactically long GS into the IPO window, 1-3 weeks, for underwriting-fee and deal-sentiment tailwind; take profit quickly because the setup is event-driven rather than fundamental.
• Buy NVDA on any post-deal pullback, 1-2 months, as the most direct beneficiary of elevated AI capex expectations; use a tight stop because the trade is dependent on continued multiple expansion.
• Relative-value pair: long NVDA / short MSFT over 2-4 months if the market starts rewarding higher-beta AI exposure over platform incumbency; this is a momentum and sentiment pair, not a fundamental call on MSFT.
• Watch for a post-IPO fade in adjacent AI infrastructure names and use call spreads instead of outright longs to cap downside; the risk/reward worsens if the market treats the SpaceX narrative as a peak enthusiasm signal.
• Avoid chasing the private-market AI complex at current levels; if the IPO is well received, wait for the first lockup expiration or a 10-15% sector pullback before adding exposure.