Anthropic Just Gave Investors One More Reason To Avoid the SpaceX IPO

SpaceX is reportedly targeting a $2 trillion IPO valuation, implying roughly 100x trailing sales on $19.3 billion of revenue, or 120x+ if measured against its post-merger scale. The article contrasts that with Anthropic’s $965 billion valuation on $47 billion of run-rate revenue and a projected $559 million Q2 operating profit, arguing Anthropic is the better-valued AI investment. The piece is primarily valuation commentary around upcoming IPOs rather than a direct operational catalyst.

Analysis

The key market signal is not that private AI is expensive; it’s that late-stage capital is now repricing duration and execution quality much more sharply than narrative breadth. A business that can show a credible path to near-term operating profitability deserves a very different multiple regime than one monetizing optionality across hardware, launch, and social/AI integration. That creates a second-order benefit for the public-market “picks and shovels” stack: as investors get more selective on model companies, demand should flow toward the compute, networking, and foundry layers where cash generation is clearer and capex is easier to underwrite.

The biggest near-term loser is the IPO window itself if pricing gaps remain this wide. A marquee debut that clears at an extreme multiple while a better-operating peer comes cheaper later can compress appetite for the whole cohort, especially among crossover funds that need relative value support. That dynamic is likely to help secondary beneficiaries like NDAQ if issuance stays hot, but it also raises the bar for every future AI float: underwriters will have to prove not just growth, but credible margin conversion within 1-2 quarters post-listing.

For NVDA and INTC, the setup is asymmetrical. Even if a premium AI company looks expensive, the implied spending still confirms sustained demand for GPUs, memory, interconnects, and advanced packaging over the next 6-12 months; the market is effectively voting to keep capex elevated. The contrarian risk is that if public-market investors push back on valuation, private buyers may slow funding pace and delay data-center orders, which would hit the supply chain before it shows up in revenue prints.

The cleanest trade is not to short the headline AI winner, but to express dispersion between monetization quality and hardware enablers. If IPO optimism reaccelerates, the winners are the names selling the picks and shovels into that wave; if it stalls, the valuation-compression trade should hurt the most levered narrative stocks first, while infrastructure suppliers remain comparatively insulated.

Key Decisions for Investors

• Long NVDA / short a basket of late-stage private AI proxies via listed surrogates or options for a 3-6 month dispersion trade; thesis is that hardware demand persists even if private-market multiples compress.
• Add to INTC on any IPO-related AI exuberance pullback over the next 2-8 weeks; if public AI valuation scrutiny forces customers to rationalize capex, Intel’s relative leverage to domestic compute demand should improve on a 6-12 month horizon.
• Use NDAQ as a volatility beneficiary: initiate a tactical long into the next 1-2 months of IPO pipeline headlines, since issuance activity and trading volumes should rise even if individual deals price conservatively.
• Avoid chasing new AI IPOs in the first 1-2 weeks post-pricing; wait for lockup or first earnings, because the setup favors headline-risk fades once valuation comparisons become tradable.
• If available, structure a call spread on NVDA rather than outright equity for the next 3-6 months; the trade benefits from continued AI capex without overpaying for a broader risk-off move in private tech multiples.