OpenAI is putting a clearer shape around a corporate shift that could redefine how one of the most closely watched AI companies is governed. The company says its current structure must evolve so it can continue pursuing its mission of ensuring artificial general intelligence, or AGI, benefits all humanity.
The plan is not simply a legal reorganization. It touches capital, control, nonprofit oversight, investor expectations and the question of who gets to decide when powerful AI systems have crossed a major threshold.
OpenAI’s proposed PBC structure
OpenAI currently operates with a for-profit organization controlled by a nonprofit. Investors and employees hold a “capped profit” share, a structure meant to limit financial upside while preserving the nonprofit mission at the top of the organization.
In a blog post published Friday, OpenAI said it intends to begin transitioning its existing for-profit into a Delaware Public Benefit Corporation, or PBC. The new entity would have ordinary shares of stock, and OpenAI’s mission would be its public benefit interest.
OpenAI framed the change as a necessary step as it enters 2025. The company wrote that it must become more than a lab and a startup, describing a world building new infrastructure around energy, land use, chips, data centers, data, AI models and AI systems for the 21st century economy.
Under the proposal, OpenAI would still have both a nonprofit and a for-profit. The PBC would run and control OpenAI’s operations and business. The nonprofit would hire a leadership team and staff to pursue charitable initiatives in sectors such as health care, education and science.
Why capital is central to the plan
OpenAI says a PBC would allow it to balance shareholder interests, stakeholder interests and a public benefit interest in decision-making. It also says the structure would let it raise the necessary capital with conventional terms.
That point matters because OpenAI’s work has become increasingly expensive. The company was founded in 2015 as a nonprofit research lab, but later created its current structure as its experiments became more capital-intensive. It took outside investments from venture capital firms and companies, including Microsoft.
In October, OpenAI raised $6.6 billion at a $157 billion valuation, bringing its total raised to $17.9 billion. Even so, the company still expects to lose money this year, with CNBC reporting a $5 billion loss expectation.
The terms of OpenAI’s latest funding round also require it to complete a for-profit transition within two years. That makes the PBC proposal more than a theoretical governance discussion. It is tied directly to the company’s financing path.
OpenAI also says the shift could create one of the best-resourced nonprofits in history. Its existing nonprofit would receive shares in the PBC “at a fair valuation determined by independent financial advisors.”
The opposition and legal pressure
The proposal faces significant hurdles. Elon Musk, one of OpenAI’s co-founders, has filed for an injunction to halt the transition. He accuses OpenAI of abandoning its original philanthropic mission.
Musk has also alleged that OpenAI deprived his AI company, xAI, of capital by extracting promises from investors not to fund it. OpenAI has called Musk’s complaints “baseless” and described them as sour grapes.
Meta, Facebook’s parent company and an AI rival, is also supporting efforts to block the conversion. In December, Meta sent a letter to California attorney general Rob Bonta arguing that the shift would have “seismic implications for Silicon Valley.”
Meta’s concern is about the precedent. In its letter, Meta wrote that if OpenAI’s new business model is valid, nonprofit investors could receive the same for-profit upside as conventional investors while also benefiting from government-granted tax write-offs.
Other AI companies are part of the comparison. Competitors such as xAI and Anthropic are structured as PBCs, but they do not have a nonprofit component.
Governance is the unresolved issue
OpenAI’s unusual structure has already had consequences. It ultimately led to the abrupt ousting of CEO Sam Altman last November, a move that displeased investors, particularly Microsoft.
The structure also gives OpenAI’s board the power to determine exactly when OpenAI has achieved AGI. That determination matters because AGI is exempt from the licensing agreements the startup has with customers.
Microsoft is one of those customers. Microsoft and OpenAI are said to have a specific, internal financial definition of AGI. According to The Information, the two companies signed an agreement last year stating that OpenAI has achieved AGI only when it develops AI systems that can generate at least $100 billion in profits.
These details make governance more than an internal operating question. They affect investors, customers, competitors and the nonprofit mission OpenAI says remains central to its identity.
Employee concerns add another layer
OpenAI is also dealing with an outflow of high-level talent, due in part to concerns that the company is prioritizing commercial products over safety.
Carroll Wainwright, a former employee who researched aligning AI systems with safety policies, wrote on X this fall that OpenAI “was structured as a nonprofit [but] acted like a for-profit” and “should not [be trusted] when it promises to do the right thing later.”
Miles Brundage, a longtime policy researcher who left OpenAI in October, also raised concerns about the transition plan. He said there was surprisingly little discussion of actual governance details, even though governance is arguably the key issue.
Brundage also argued that a well-capitalized nonprofit on the side is not a substitute for PBC product decisions being aligned with the original nonprofit mission. He questioned whether charitable work in health care, education and science is too narrow compared with OpenAI’s original mission, and asked what happens to safety and good policy.
His broader concern is that the nonprofit could become a “side thing” while the PBC operates like a normal company. That is the central tension in OpenAI’s proposal: whether a public benefit corporation and a wealthy nonprofit can preserve the mission OpenAI was founded to pursue while giving the company the capital structure it says it now needs.