A Boardroom in Flux: The OpenAI Leadership Crisis
When Sam Altman was briefly ousted as OpenAI’s CEO in November 2023, the turmoil was more than a personality clash. It triggered a rapid scramble to redefine control over one of the most influential AI labs in the world. Newly unredacted trial exhibits from Musk v. Altman show that, behind the scenes, Microsoft executives became deeply involved in shaping a new OpenAI board. The OpenAI leadership crisis unfolded just weeks after a highly public show of unity between Satya Nadella and Altman on stage at OpenAI Dev Day. Yet within days of Altman’s removal, text threads reveal Nadella, Microsoft CTO Kevin Scott, and president Brad Smith assessing potential directors while Altman himself asked for their input. Those exchanges now provide an unusually clear view of how a major strategic investor can navigate, and subtly steer, AI company governance during a moment of existential uncertainty.
Satya Nadella’s ‘Next IBM’ Warning and the Stakes for Microsoft
Unsealed emails presented at trial show why Microsoft was so invested in OpenAI’s future. In an internal message from April 2022, Satya Nadella warned that he did not want Microsoft to become “the next IBM” while OpenAI became “the next Microsoft.” He described the OpenAI partnership as a “one-way door,” since Microsoft could not feasibly build two separate supercomputers and had to channel scarce computing resources into OpenAI’s infrastructure. On the stand, Nadella said the company was “outsourcing essentially a lot of the core IP development,” creating a massive dependency on OpenAI’s research pipeline. That dependency heightened Microsoft’s need for visibility into, and confidence in, OpenAI’s leadership and board structure. Nadella insisted he sought access to OpenAI’s intellectual property and the ability to build internal expertise in parallel, framing the relationship as mutual risk-taking rather than passive investment.
Vetting the Board: Who Microsoft Backed—and Blocked
The text messages between Nadella, Kevin Scott, Brad Smith and Sam Altman reveal a meticulous, strategic vetting of candidates for a revamped Microsoft OpenAI board. Scott floated former Google Cloud CEO Diane Greene, prompting a “strong, strong no” from Scott himself and objections from Nadella due to her ties to a key competitor in AI. Nadella also opposed veteran gaming executive and long-time Amazon board member William “Bing” Gordon, again citing competitive conflicts. In contrast, Belinda Johnson, the former Airbnb COO, drew enthusiastic praise from Scott, while Nadella proposed former Gates Foundation CEO Sue Desmond-Hellmann, who ultimately joined the OpenAI board. Other names ranged from Ursula Burns and Anne Sweeney to Leslie Kilgore and tech and finance veterans such as Amy Rao, Emilie Choi and Jeff Weiner. Even Scott jokingly offered himself, a suggestion Nadella quickly dismissed, underscoring Microsoft’s sensitivity to perceived overreach.
From Text Thread to Boardroom: Navigating Risk and Influence
By the end of that frenzied day, a provisional framework had emerged: a three-person board of Bret Taylor, Larry Summers and Adam D’Angelo, with Sam Altman reinstated as CEO but kept off the board. Brad Smith warned that Summers, despite his intellect, was “so mercurial,” calling his appointment risky. Altman conceded the concern yet argued he would accept it to stabilize OpenAI, asking if Microsoft could “live with it.” Nadella replied by requesting to call Summers directly, a move that highlights how closely Microsoft tracked each board choice. Nadella later testified that OpenAI’s board was free to ignore his suggestions and emphasized that Altman and other insiders initiated the outreach. Still, the final board composition—eventually including Desmond-Hellmann and Taylor—shows how an investor’s preferences can align with outcomes, even without formal control, especially during a high-stakes leadership reset.
AI Company Governance on Trial
Elon Musk’s lawsuit casts Microsoft’s role as a betrayal of OpenAI’s nonprofit mission, accusing the company of aiding a breach of the charitable trust underpinning the lab’s founding. His attorneys argue that Microsoft’s efforts to safeguard its commercial stake came at the expense of OpenAI’s commitment to develop AI for the benefit of humanity. Nadella countered in court that Microsoft took “enormous risk” to support a lab that no one else would fund, helping create one of the largest nonprofits in the world and enabling products like ChatGPT and Copilot. Under cross-examination, he acknowledged he was unaware of any full-time staff at the nonprofit before March 2026 or any grants or open-source work it had produced. The clash underscores a broader tension in AI company governance: how to balance mission-driven structures with the realities of capital-intensive partnerships and powerful corporate stakeholders.