SpaceX’s record listing is done. Two more stand behind it in the queue, and the combined scale has no precedent. OpenAI is aiming at a $1 trillion valuation, and Anthropic is marked at $965 billion in the private market, which puts as much as $2.5 trillion of newly listed value into a six-month window, according to an analysis in The New York Times. Three of the largest public debuts in history, arriving at once, is a concentration of risk that even veteran bankers say they have not seen.
The historical record is uncomfortable. Giant IPOs have a habit of arriving near market tops. Uber went public in May 2019 at $45 a share, closed its first day below the offer price and spent the following years below it. AT&T Wireless raised more than $10 billion in an April 2000 offering, weeks before the technology bubble broke. Rivian, whose 2021 IPO was then the largest of its kind, has traded far below its listing price for years. Facebook, whose 2012 debut was the biggest technology listing of its era, saw its stock fall below the $38 offer price within months. The pattern, analysts said, is not coincidence: the conditions that make investors willing to pay for a massive offering, easy money and euphoria, are the same conditions that precede the reckoning.
The current wave stacks those precedents on top of one another. SpaceX listed at a valuation above $2 trillion and raised $75 billion, the largest sum any company has ever taken from public investors. OpenAI, if it follows at the $1 trillion mark, would roughly match the size of the deal. Anthropic at $965 billion would be close behind. Three companies, none of which existed as public entities at the start of the decade, would collectively absorb an amount of investor capital that exceeds the annual gross domestic product of most countries.
The bull case is not hard to construct, and the companies’ backers make it loudly. The AI trade has actual revenue behind it this time, unlike the dot-coms that burned through cash without a product. Nvidia’s earnings, the data-center buildout and enterprise spending on models all suggest that the demand driving these valuations is real. The AI companies themselves are growing revenue at a pace few public companies have matched, and their largest customers are the richest companies in the world, not speculative day traders. Retail demand is also a new factor: the same platforms that popularized single-stock trading have made it possible for millions of small investors to participate in a giant offering on the first day, a source of buying that did not exist when Uber and Facebook listed. SpaceX, for its part, has a monopoly-grade launch business and a satellite network with recurring revenue. The argument that this is a new industrial cycle, not a bubble, has the evidence of the earnings reports behind it.
The bear case rests on the mechanics of supply. A wave of giant offerings floods the market with shares that must be absorbed by the same pools of capital, and each successive deal competes with the last. The window is unusually compressed: three enormous listings within about six months means the lockups, the shareholder sales and the index inclusion events all arrive in the same period. When the marginal buyer is exhausted, the tape turns, and the stocks most recently bid up are the ones that fall hardest.
There is also a structural difference from past waves that cuts both ways. Companies now stay private far longer, which means the public market no longer sees them young and cheap; it sees them at full size. That concentrates both the upside and the downside of a listing in a shorter window, and it raises the cost of getting the timing wrong. The passive-investing machine, meanwhile, guarantees a floor of demand for any giant that enters an index, which supports prices in the short run but leaves a large pool of owners who bought because the rules told them to, not because they judged the stock.
There is also the question of what the mega-IPO wave says about the cycle’s position. Companies time their listings to the moment of maximum demand, and the smartest money in the private market tends to sell into strength. The fact that the three largest private companies in technology all chose the same window suggests that their owners look at the same valuation peak everyone else sees. A founder selling at the top is not proof of a top, but it is not a contrary indicator either.
The New York Times analysis notes that the 2019 Uber example is the closest parallel to the current moment, and the lesson cuts both ways. Uber’s listing marked the end of a long period of easy money for money-losing companies, and its stock did not recover for years. But the companies that survived the reckoning, including Uber, eventually built durable businesses. The question for the AI trio is not whether the boom ends, but whether the businesses underneath the boom survive the end. The answer will determine whether $2.5 trillion of listings is remembered as a peak or a beginning.
