Musk v. Altman week 2: OpenAI fires back, and Shivon Zilis reveals that Musk tried to poach Sam Altman

The legal confrontation between Elon Musk and OpenAI has entered a volatile second week, shifting from abstract philosophical debates about artificial intelligence safety to a gritty evidentiary battle over personal motivations and corporate betrayal. Musk’s lawsuit, which centers on the allegation that OpenAI abandoned its non-profit roots for a lucrative partnership with Microsoft, hit a significant snag with the testimony of Shivon Zilis. A high-ranking executive at Neuralink and mother to three of Musk’s children, Zilis revealed that Musk once attempted to poach Sam Altman for a competing project, complicating Musk’s narrative that he was a purely altruistic benefactor deceived by profit-hungry entrepreneurs.

This litigation resides at the intersection of a decade-long power struggle for the future of AGI. When OpenAI was founded in 2015, Musk, Altman, and Greg Brockman presented a unified front against Google’s perceived monopoly on AI talent. Musk’s current stance is rooted in the "Founding Agreement"—a document OpenAI claims does not formally exist—which supposedly mandated that the lab’s technology remain open-source and for the benefit of humanity. The context of their eventual fallout in 2018 is now being litigated in real-time, as the court examines whether Musk’s exit was due to legitimate safety concerns or a failed attempt to merge OpenAI into Tesla.

The technical and business mechanics under scrutiny revolve around OpenAI’s transition to a "capped-profit" model. Musk’s legal team argues that the $13 billion partnership with Microsoft effectively transformed OpenAI into a closed-source subsidiary of the tech giant. OpenAI, conversely, argues that the massive compute requirements for modern LLMs necessitated a capital structure that traditional non-profits simply cannot support. The court is tasked with deciphering whether the shift was a pragmatic evolution required for survival or a breach of fiduciary duty to the original donors who backed a public-interest mission.

Industry implications of this trial are profound, as the verdict could redefine how non-profit governance interacts with commercial scaling in the tech sector. If Musk succeeds, it could set a precedent that restricts the ability of AI labs to privatize breakthroughs derived from initial non-profit research. This has caused jitters among venture capitalists and board members who favor the "dual-structure" model. Furthermore, if OpenAI is compelled to open-source its most advanced models, the competitive landscape would flip overnight, potentially eroding the proprietary moat currently enjoyed by Microsoft and OpenAI.

The interpersonal drama revealed by Zilis adds a layer of reputational risk that transcends the courtroom. Evidence suggesting that Musk sought to lure Altman away to his own ventures undermines his claim that his primary goal was protecting OpenAI’s mission. It paints a picture of a competitive landscape where “safety” and “altruism” are often used as rhetorical shields for standard talent wars and market positioning. This development suggests that the trial may devolve into a "he-said, she-said" regarding private conversations and intent, rather than a clear-cut interpretation of contract law.

Moving forward, the tech world should watch for the discovery phase to unearth further internal communications from the 2015–2018 era. These emails will likely clarify whether OpenAI’s founders always intended to pivot toward profit or if the decision was a reactive move to the escalating costs of the GPU arms race. Additionally, the role of federal regulators remains a wildcard; if the trial exposes systemic gaps in non-profit oversight, we may see fresh legislative pushes to codify how "human-centric" AI labs must operate. For now, the rift between Musk and Altman stands as the definitive schism of the AI age.