OpenAI completes shift to Public Benefit Corporation: what changes for those building on the API
Completed in October 2025, OpenAI's recapitalization swapped the profit cap for conventional shares and created a Public Benefit Corporation. For those betting their entire product on the API, the governance change redraws the risk around pricing power and platform dependency.
From nonprofit research to a structure with real equity
OpenAI was founded in 2015 as a nonprofit organization, with the stated mission of ensuring that artificial general intelligence benefits all of humanity. In 2019, to fund the scale of its research, the entity created a for-profit subsidiary, but one with limited profit (capped-profit), always controlled by the nonprofit organization.
On October 28, 2025, according to the official Our Structure announcement published on OpenAI's website, this architecture was reworked. The nonprofit entity became OpenAI Foundation. The commercial subsidiary became OpenAI Group PBC, a Public Benefit Corporation: a corporate form that, by law, must advance its stated mission and consider the interests of all stakeholders, not just shareholders.
The recapitalization was not an informal adjustment. According to OpenAI itself, it was completed after nearly a year of dialogue with the attorneys general of California and Delaware, the bodies that oversee this type of corporate restructuring in the US. This signals that the final design went through specific regulatory scrutiny over the risk of the original mission losing weight against capital.
What didn't change: the Foundation still calls the shots
OpenAI Foundation continues to control OpenAI Group. Through special voting and governance rights, it is the Foundation that appoints the entire Group board and can replace directors at any time. The Safety and Security Committee, responsible for overseeing safety practices across the whole operation, remains a Foundation committee, not a Group one.
The Foundation board includes names such as Bret Taylor (chair), Adam D'Angelo, Paul Christiano, Sue Desmond-Hellmann, Zico Kolter, retired general Paul Nakasone, Adebayo Ogunlesi, Nicole Seligman, Robin Vince and CEO Sam Altman. For those following the Brazilian ecosystem, one name stands out: David Vélez, founder and CEO of Nubank (a Brazilian digital bank), also sits on the Foundation board.
Vélez's presence doesn't mean preferential treatment for Brazilian startups, but it is a concrete indicator that the board that sets OpenAI's direction does have a Brazilian with fintech scaling experience at the decision-making table.
What disappeared: the investor return cap
The central financial change in the recapitalization lies in a technical detail that shifts the long-term incentive across the entire capital chain behind OpenAI.
| Aspect | 2019 structure | Post-recapitalization structure |
|---|---|---|
| Board control | Nonprofit controlled the for-profit | Foundation continues to control and appoint the Group board |
| Investor return | Limited profit (capped-profit) | Conventional common stock, no cap, proportional stake |
| Legal nature of the commercial arm | Capped-profit LLC | Public Benefit Corporation |
| Controlling entity's stake | No traditional equity | Foundation holds 26% of the Group (~US$130 billion) plus a 15-year upside warrant |
The announcement itself confirms: "all equity holders in OpenAI Group now hold the same type of traditional stock, which participates proportionally and grows in value with OpenAI Group's success." This removes the brake that the 2019 model placed on investor returns.
Doing the math with OpenAI's own numbers, the Foundation's 26% stake is worth about US$130 billion, which implies a total valuation for OpenAI Group of around US$500 billion. This figure isn't stated in the announcement; it's a direct inference from the percentage and value provided.
Why this matters for those who built products on top of the API
For a Brazilian founder whose product depends entirely on OpenAI's API, swapping capped-profit for conventional, uncapped equity changes the risk calculus in two practical ways.
First, pricing power. Without the return cap that, in theory, limited how much value the commercial structure needed to extract, the long-term incentive for investors and the board becomes maximizing revenue growth with no formal limit. This doesn't mean a price hike tomorrow, but it removes a structural barrier that was historically cited as a counterweight to aggressive API monetization.
Second, platform dependency. A company valued at around US$500 billion, under pressure to deliver returns proportional to that valuation, has a stronger commercial motive to push corporate upsell, tighten usage limits on smaller plans, or create pricing tiers that favor those already locked into the ecosystem. Those who built their entire stack on gpt-* with no plan B feel this shift first.
The counterpoint: a PBC is not a mandate for pure profit
The pricing power risk thesis has a structural counterargument that is also present in OpenAI's own document, and it needs to be taken seriously before any alarmist conclusion.
which—unlike a conventional corporation—is required to advance its stated mission and consider the broader interests of all stakeholders, ensuring the company's mission and commercial success advance together.
OpenAI, official "Our Structure" announcement
A Public Benefit Corporation has a legal obligation, albeit a loose one in practice, to weigh impact beyond shareholder return, unlike the classic fiduciary duty to maximize shareholder value that governs most publicly traded companies in the US. Add to this the fact that the Foundation, not investors, appoints and replaces the Group board, and has its own safety committee with oversight power over the entire commercial operation.
In practice, the design tries to shield exactly the risk that product founders fear most: pricing decisions driven solely by pressure for financial returns. Whether this will hold up once the capital raised needs to be paid back at scale is a question neither party, OpenAI nor its API customers, can answer today with concrete data.
What changes in practice for those deciding where to allocate product and money
For those who already decided to build their business's core on top of OpenAI's API, the October 2025 recapitalization is not a reason for an immediate exodus, but it is a concrete reason to review two points of risk: long-term contracts without a predictable repricing clause, and product architecture without an abstraction layer that allows switching model providers without rewriting the entire application.
The most useful signal for a Brazilian founder isn't the US$500 billion valuation itself, it's the incentive shift behind it: a company that no longer has a formal return cap has a stronger internal commercial argument to monetize its installed base. Anyone depending 100% on a single model provider is betting that this argument will never outweigh the mission stated on paper.
Translated from the Brazilian Portuguese original · Read the original
OpenAI formalizes the shift to PBC and redraws who controls the API
Since October 2025, the structure underpinning OpenAI stopped being 'capped-profit' to become a Public Benefit Corporation controlled by a foundation. For those building products on top of the API or negotiating partnerships with the company, this redraws the return cap, governance, and the risk of depending on a single vendor.
