SpaceX, OpenAI, and Anthropic Are Racing Toward a $4 Trillion IPO Wave — But Is It a Bubble?

2. A First-Mover Race Is Underway

Whoever lists first sets the pricing benchmark for the rest. SpaceX, the farthest along, began its roadshow on June 5 and is expected to price as early as June 12 . OpenAI filed its confidential S-1 on May 22, targeting a September–Q4 2026 debut . Anthropic leapfrogged its rival by filing on June 1, just days after closing a $65 billion round that valued it at $965 billion — temporarily surpassing OpenAI's $852 billion private valuation . Each company is trying to get out the door before a competitor's poor post-IPO performance sours the market.

3. The Exit Window May Be Closing

With AI stocks already making up roughly 35% of the S&P 500, investor enthusiasm is at a peak . Bank of America's Michael Hartnett has noted that the top 10 AI stocks account for around 40% of total market capitalization, and the coming IPOs would push that group toward 50% — a level of concentration never seen before . The private backers who rode these valuations to nearly a trillion dollars each want to cash out before any potential correction.

The Numbers: What Each Company Is Targeting

SpaceX confidentially filed its S-1 on April 1, 2026, and released it publicly on May 20. It is targeting a Nasdaq listing under ticker SPCX, with a $1.75–$2.0 trillion valuation and a $75–$80 billion raise, making it the largest IPO in history by a wide margin . The roadshow is underway, and pricing is expected between June 18 and 30, with a possible debut as soon as June 12 . The S-1 revealed that Starlink — SpaceX's satellite internet business — is the financial engine. It generated $11.39 billion in revenue in 2025 (61% of SpaceX's total, up nearly 50% year-on-year), posted an operating profit of $4.42 billion, and carried an EBITDA margin of 63% with 10.3 million subscribers . Q1 2026 alone saw Starlink produce $3.26 billion in revenue and $1.19 billion in operating income .

OpenAI filed its confidential S-1 on May 22, 2026, with Goldman Sachs and Morgan Stanley underwriting . The target valuation band is $850 billion to $1 trillion, against a last private round of $852 billion in March 2026 . The company is aiming for a public debut as early as September 2026 . It has approximately $25 billion in annualized revenue, but also roughly $14 billion in annual losses, burning cash heavily on compute infrastructure . One analysis suggested OpenAI needs an estimated $207 billion in additional capital through 2030 just to honor its existing compute commitments — making the IPO less a victory lap than a forced funding event .

Anthropic filed its confidential S-1 on June 1, 2026, after a $65 billion Series H that valued the company at roughly $965 billion . Its revenue run-rate has exploded from $1 billion to over $19 billion as of February 2026, and it has secured up to $40 billion from Google and $25 billion from Amazon . No IPO pricing date has been set, but the company is targeting a fall 2026 debut . Crucially, the widely-cited "$965 billion" figure is a private valuation, not a public IPO target — and some analysts estimate the actual IPO could price at a significant discount, potentially between $400 billion and $600 billion .

The Bull Case vs. The Bear Case

Expert opinion is split between two starkly different narratives.

The Bull Case: This Is Not Pets.com

Proponents argue that the 1999 dot-com analogy is misleading because these companies have real, often profitable, revenue . Unlike the dot-com era, when roughly 80% of IPOs had no profits, SpaceX's Starlink is a genuine cash machine . In Q1 2026 alone, Starlink generated more operating income ($1.19 billion) than the entire launch business produced in revenue . Anthropic's revenue has grown nearly 20× in about a year, from $1 billion to $19 billion run-rate . OpenAI has a globally recognized brand, hundreds of millions of users, and $25 billion in annualized revenue .

Bulls also argue that the concentration in AI stocks is a reflection of genuine structural shifts — not just speculative froth. As one analysis summarized, the dot-com analogy "is most useful as a warning about the sequence of events — capex peak → utilization disappointment → guidance cuts → cascade — rather than as a 1:1 template" .

The Bear Case: The Multiples Are Historically Extreme

The revenue multiples at which these companies are being valued are staggering, even by the standards of high-growth tech.

• SpaceX — At $1.75 trillion and roughly $18.7 billion in 2025 revenue, it is trading at approximately 94× trailing revenue . By comparison, even Nvidia, Tesla, and Amazon at comparable stages never approached these levels. On a forward basis using 2026 projections of $28.5 billion, the multiple is still roughly 61× .
• Anthropic — At its $965 billion Series H pricing, the company trades at about 21× current annualized revenue run-rate .
• OpenAI — At $852 billion, the company prices at a wide multiple of trailing revenue while remaining deeply unprofitable, with losses of roughly $14 billion .

The most prominent institutional bear is Morningstar. The independent research firm published a fair value estimate of just $780 billion for SpaceX — less than half the $1.75 trillion IPO target — and warned investors that they "will have opportunities to buy the stock at more attractive levels" after the offering . Morningstar equity analyst Nicolas Owens anchored the valuation to SpaceX's launch and Starlink businesses and applied probability weighting to the AI segment's various development scenarios .

Other analysts have reached similar conclusions using different models. One sum-of-the-parts analysis pegged SpaceX's median fair value near $1.25 trillion, roughly 30% below the IPO target . Another projected that even if SpaceX reaches $50 billion in revenue by 2029 and trades at a generous 35× multiple, the implied market cap would be exactly $1.75 trillion — meaning the IPO is pricing in nearly a decade of flawless execution today .

The Dot-Com Question: Bubble Signal or Rational Pricing?

The dot-com comparisons are not just rhetorical. In inflation-adjusted terms, the combined value of these three IPOs would exceed the valuation of every U.S. IPO from 1995 to 2000 combined . A Reuters analysis noted that the three companies are, as a group, losing money — a combination without precedent in U.S. market debuts at this scale .

Yet the comparison is imperfect. The AI cycle has so far "not produced the negative-earnings IPO frenzy that characterized late-stage dot-com mania," one analysis noted . Starlink is profitable and fast-growing. The underlying technologies — large language models, reusable rockets, satellite broadband — have demonstrable utility and adoption that the concept stocks of 1999 lacked .

The more precise risk is not that these companies are worthless, but that they are worth significantly less than their IPO prices suggest. The gap between private secondary-market euphoria and what public-market institutional investors are willing to pay is the central uncertainty. Some analysts predict Anthropic's public valuation could land 40–60% below its $965 billion private round . If the first of these IPOs trades poorly, it could create a cascade that reprices the entire AI capex cycle.

The Market Absorption Problem

The timing itself is a risk factor. SpaceX is seeking $75–$80 billion in June . OpenAI could seek $60 billion or more in September . Anthropic's raise size is not yet set but will likely be in the tens of billions. Even in a liquid market, absorbing over $200 billion in new issuance from a single sector in a matter of months would be historically unprecedented .

Institutional investors with existing concentrated positions in large-cap tech may be forced to sell existing holdings to make room for these new listings, creating mechanical selling pressure across the market . The later filers — OpenAI and Anthropic — could face significantly worse pricing if SpaceX's debut is rocky .

What to Watch Next

The most important data points that will resolve the bull/bear debate are still ahead. For SpaceX, all eyes are on the IPO pricing on or around June 12 and how the stock trades in its first weeks . For OpenAI and Anthropic, the public release of their confidential S-1 filings — expected roughly 21 days before their respective roadshows — will provide the first definitive look at revenue, gross margins, profitability, and risk factors .

The question is not whether AI and commercial space are genuinely transformative industries. The question is whether public-market investors, staring at 94× revenue multiples and $14 billion annual losses, will pay the same prices that private-market investors paid in the euphoric final months before the window opened — or whether this historic clustering of mega-IPOs will mark the moment the market drew a line.