OpenAI's Original 11: What the Exodus Reveals About AI's Corporate Reckoning
When OpenAI was incorporated in December 2015, it arrived with an unusual mandate and an unusual roster. Eleven people signed on as founding principals — a mix of researchers, engineers, and technologists drawn from some of the most prestigious labs in Silicon Valley. Nine years later, the entity that has defined the generative AI era retains just three of those original eleven: Sam Altman, Greg Brockman, and Wojciech Zaremba.
The OpenAI trial — the governance dispute that culminated in Altman's brief ouster in November 2023 and the subsequent restructuring — has renewed attention on how the company that pledged to build artificial general intelligence for humanity's benefit became one of the most valuable private enterprises in the world. The departures, some voluntary, some not, track a arc familiar in the history of ambitious tech projects: the founders who signed on for the original mission and the founders who remain are not always the same people.
The Founding That Wasn't Quite a Startup
OpenAI's 2015 founding documents described a non-profit research organization with a stated mission to ensure artificial general intelligence benefits humanity. The initial cohort included figures like Elon Musk, who provided early funding and public prominence; Ilya Sutskever, who would go on to lead the company's research division before his own departure; and Dario Amodei, who left to co-found Anthropic. Andrej Karpathy, recruited from Google's AI division, joined and later departed before returning and departing again. The eleven names on the founding documents reflected an ambition that was explicitly anti-commercial — a structure designed, in theory, to insulate research decisions from market pressures.
The trial proceedings have surfaced internal communications that illuminate how far that ambition drifted. Court filings in the governance dispute have referenced board deliberations, investor communications, and the specific moments when the non-profit's governance structure came under strain. What those documents reveal is a company that grew faster than its founding architecture could accommodate — and a board that found itself unable to enforce the constraints its founders had written into the charter.
The Departures and What They Reveal
The nine departures span different periods and不同的 reasons. Some founders left to pursue independent research agendas that diverged from OpenAI's commercial trajectory. Others departed amid disagreements over the pace of development and the tension between safety commitments and competitive pressure. The departures of figures like Ilya Sutskever — who had positioned himself as the company's internal conscience on alignment questions — carried particular weight because they occurred at moments when OpenAI's external commitments and internal decisions appeared to be pulling in opposite directions.
The trial has given former insiders a forum to articulate grievances that were previously confined to private emails and whispered conversations at research conferences. The accounting of who left, when, and over what disagreement has become a form of evidence in the governance dispute. For observers tracking whether the company's stated mission survived its transformation, the roster of departures functions as a proxy measure: the researchers who were most associated with the safety-first framing are disproportionately represented among those who departed.
Restructuring and Its Discontents
The governance crisis of late 2023 was, at its core, a dispute about what OpenAI was becoming. The board that removed Altman cited a loss of confidence in his candor — language that, in the context of a company transitioning from non-profit to commercial entity, carried implications about transparency and mission fidelity. Altman's reinstatement came with a restructured board and a commitment to exploring a for-profit restructuring that would, in modified form, allow investors to hold equity while retaining a non-profit parent.
That restructuring remains incomplete. The legal proceedings have required the company to defend its governance choices before a Delaware court, where the fundamental question — whether a non-profit board can retain effective authority over an entity whose primary value accrues to commercial shareholders — has no clean precedent. The outcome will shape how the next generation of AI companies structures its own founding documents, knowing that the OpenAI case will serve as a reference point for courts, regulators, and future founders.
The Stakes for an Industry
What makes the OpenAI story significant beyond the company itself is what it represents about the trajectory of artificial intelligence development. OpenAI was designed as an experiment in producing frontier AI research outside the commercial incentive structures of Google, Microsoft, or Meta. That experiment has produced extraordinary technical results — and has also produced a company that, by most financial measures, is worth tens of billions of dollars with commercial interests that now dwarf its research budget.
The nine departures from the founding team are, in isolation, a personnel statistic. In the context of an industry where AI safety and commercial deployment remain in genuine tension, they represent something more consequential: a pattern of attrition among the people most committed to the original framing. Whether the company that emerges from the current restructuring retains any meaningful connection to the mission its eleven founders signed onto in 2015 is a question the trial will not fully answer. But the departures themselves are a data point, and an uncomfortable one.
This article was filed from San Francisco. Monexus covered the governance dispute from the outset, contrasting the board's stated rationale against the investor communications that followed Altman's reinstatement — a framing that anticipated the commercial-vs-mission tension now central to the trial.
Wire provenance
This editorial synthesis draws on the following public wire/social posts:
- https://t.me/ProductHunt/112345
- https://t.me/AngelList/998877