Two arch-rivals are heading for the public markets through the same two banks — and the arrangement is already testing how Wall Street polices its own conflicts of interest.
By the numbers
Two of the most anticipated stock-market debuts in years are taking shape — and before a single share has changed hands, they are already testing how Wall Street polices its own conflicts of interest. According to reporting from The Wall Street Journal, Goldman Sachs and Morgan Stanley, the two most prestigious names in equity underwriting, will jointly lead the initial public offerings of both OpenAI and Anthropic, the two leading artificial-intelligence labs and direct competitors for customers, engineers and capital.
What an IPO actually is
When a private company “goes public,” it sells shares to outside investors on a stock exchange for the first time. It is the moment early backers and employees can finally turn paper wealth into cash, and the moment the wider public — pension funds, mutual funds and everyday investors — gets the chance to own a slice. Companies rarely attempt this alone. They hire investment banks as underwriters, and those banks do the unglamorous, high-stakes work: gauging how much investors will pay, setting the price, lining up the large institutions that will anchor the deal, and effectively staking their own reputations on it. The most senior banks on a deal are called the “bookrunners,” and they sit at the centre of the entire process — which is precisely why a lead role on a marquee IPO is among the most coveted assignments in finance, worth both enormous fees and lasting prestige.
Same banks, two rivals
Here is the twist. Goldman and Morgan Stanley are not leading one of these landmark offerings — they are leading both, for two companies locked in direct competition. That arrangement creates an obvious tension: the same institutions could end up holding the most sensitive financial details of two arch-rivals at the very same moment, from revenue and profit margins to growth plans and strategy. To manage the risk, the Journal reports that each bank is assembling separate, “bespoke” deal teams, so that information shared by one company never reaches the people advising the other.
How one bank serves two rivals
Confidential information can’t cross between the two teams.
The wall Wall Street runs on
Wall Street has a name for this kind of internal divider: a “Chinese wall,” or information barrier. It is a combination of rules, training and physical and digital separation designed to stop confidential knowledge in one corner of a bank from leaking into another. Such barriers are routine — banks advise competing clients all the time — but running two rival, blockbuster technology IPOs through the same two firms at once is an unusually demanding stress test of whether those walls truly hold. The stakes run in both directions. If either company suspects its secrets are not safe, it can take its business elsewhere; and if investors come to believe the process is compromised, confidence in the pricing of the shares — the very thing the banks are paid to get right — begins to erode.
Before a single share trades, these IPOs are already a case study in how the same few institutions serve fierce competitors at once.
Why it matters for you
It is tempting to file all this under “billionaire problems,” but the ripples reach much further. These are expected to be among the largest and most closely watched offerings in recent memory, and investor appetite for them will be read as a referendum on the entire AI investment boom. Strong debuts would signal that public markets still believe the economics of artificial intelligence justify the enormous sums being poured into it; weak ones would raise uncomfortable questions about valuations across the sector. For ordinary savers, that signal travels — into tech-heavy index funds, into the retirement accounts that hold them, and into the broader mood of the market — well beyond anyone who ever buys a single share.
A frontier-tech IPO wave
There is also a striking marker of how quickly this landscape is shifting. Sequoia Capital’s Roelof Botha recently recalled predicting a year ago that SpaceX could one day be worth more than OpenAI — and noted that it has now happened, with SpaceX already trading publicly. The frontier-tech names that dominated private fundraising are now marching toward the public markets one after another, and the OpenAI–Anthropic pairing is simply the headline act.
What to watch
A few signals will tell the story as it unfolds. The first is valuation: the price each company targets will be a direct read on how investors rate AI leadership, and any gap between the two will be picked over endlessly. The second is the syndicate — whether more banks are invited to join the deal, which would spread both the fees and the risk and dilute the awkwardness of two rivals sharing the same advisers. The third is timing: two giant, competing IPOs chasing the same pool of investor money in a narrow window may not sit comfortably together, and one could be pushed to move first while the other waits for a friendlier moment.
Key takeaways
- The same two banks, Goldman Sachs and Morgan Stanley, are leading the IPOs of direct rivals OpenAI and Anthropic.
- To prevent leaks, each bank is building separate, walled-off deal teams — a classic “information barrier” now under unusual strain.
- The debuts will be read as a verdict on the whole AI investment boom, with ripple effects for ordinary investors’ funds.
Source: Axios Pro Rata (Dan Primack), June 18, 2026, citing The Wall Street Journal.