The ongoing courtroom drama between Elon Musk and Sam Altman over the future of OpenAI has taken a new turn, with key testimony shedding light on the inner workings of the artificial intelligence research organization. During a recent hearing, a witness named Eddy stated, "The documents tell the truth here," emphasizing that the written records from early brainstorming sessions do not support the concept of an "adjunct" for-profit entity—a term Musk used during his testimony. The observation aligns with what many analysts have been following: the legal battle revolves around the fundamental question of whether OpenAI, originally founded as a nonprofit, was always intended to remain one, or if a for-profit conversion was always part of the plan.
The phrase "adjunct" for-profit was a focal point of the court proceedings. When Musk used it in his testimony to describe a potential structure, it was met with objections from Altman's legal team, led by attorney Savitt. The objection accused Musk's camp of inventing a term not present in any of the founding documents. Eddy, whose role in the case appears tied to document review, reaffirmed that searching through early brainstorming documents reveals no mention of an "adjunct" entity. Instead, documents show parallel for-profit organizations and explicit discussions about converting the entire venture to a for-profit model while shutting down the nonprofit side. This distinction could prove critical in determining whether Musk's allegations of a broken promise have legal standing.
Further complicating matters, Jared Birchall, a key figure in Musk's orbit, testified that he had filed paperwork to register a company for this exact purpose. Birchall, who has been involved in various Musk ventures including Tesla and SpaceX, stated that his actions were in line with what Musk was proposing at the time. However, the lack of a clear paper trail matching Musk's narrative has left the court to weigh the credibility of oral testimony against the written record that Eddy champions. Judge presiding over the case has yet to issue a ruling, but the testimony highlights the intricate corporate maneuvering behind one of the most high-profile AI startups in the world.
OpenAI was established in 2015 as a nonprofit research company with the mission of developing artificial general intelligence (AGI) that benefits humanity. Initially, it was funded by a group that included Musk, Altman, and other tech luminaries. Musk donated heavily but stepped down from the board in 2018 to avoid conflicts with Tesla's AI work. In 2019, the organization launched a capped-profit subsidiary called OpenAI LP to attract outside investment, a move that Musk later criticized as a betrayal of the original nonprofit ethos. Altman, now CEO, has defended the shift as necessary to compete with deep-pocketed rivals like Google and Microsoft, which have invested billions in the company.
The lawsuit, filed by Musk in early 2024, alleges that Altman and the current board breached their fiduciary duties by pursuing commercial interests over the public good. Musk's team argues that the capped-profit structure was a stepping stone to full privatization, leaving the nonprofit arm largely defunct. The defense counters that the structure was always transparent and that investors like Microsoft were aware of the dual-purpose model. The "adjunct" for-profit testimony adds a layer of nuance: if there was no plan for a secondary, non-profit-linked entity, then Musk's claim that OpenAI deceived donors and researchers may be weakened.
Industry observers note that this case could set a precedent for how charitable organizations can pivot to for-profit models, especially in the fast-moving AI sector. The outcome may influence other AI labs, such as Anthropic and Cohere, that have also adopted hybrid structures. Moreover, the case touches on the broader ethical debate about governance in AI companies. As the technology's power grows, questions about transparency, accountability, and alignment of interests become paramount. The testimony from Eddy and Birchall has brought these issues into sharper focus, but the ultimate resolution depends on more evidence and legal arguments.
Additional details from court filings reveal that early discussions among OpenAI's founders included multiple scenarios. One proposed structure involved creating a for-profit arm that would be "parallel" to the nonprofit but with strictly separate leadership. Another option was a full conversion, with the nonprofit either dissolved or repurposed to fund philanthropic AI safety research. The documents show that neither scenario explicitly used the term "adjunct," but the parallel-for-profit concept came close. However, the distinguishing factor is that a parallel structure maintains the nonprofit's independence, while an adjunct suggests subordination. Musk's testimony implied OpenAI exploited the adjunct label to reassure regulators that the nonprofit remained central, even as the for-profit grew dominant.
The judge has yet to rule on motions for summary judgment, but legal experts predict the case could go to trial later this year. If it does, the document trail will be a central piece of evidence. Eddy's assertion that "the documents tell the truth" reflects a philosophy that written records are more reliable than memory, especially when the individuals involved have conflicting recollections. The court has ordered both sides to produce additional emails, notes, and board minutes from 2015 to 2020, which could further clarify the original intent.
Beyond the immediate lawsuit, the case has become a flashpoint in the broader debate about AI safety and profit incentives. Critics argue that OpenAI's shift from a purely nonprofit model has compromised its commitment to cautious development, while supporters maintain that financial resources are necessary to conduct the most rigorous safety research. The court's analysis of the founding documents might influence not only legal judgments but also public perception of the organization's integrity.
As the hearing continued, testimony turned to the role of key individuals. Baron, another witness involved, is expected to provide insight into the early culture of OpenAI. The company originally styled itself as a transparent, collaborative community that would share its research freely. Over time, however, competitive pressures led to increased secrecy and a focus on proprietary models like GPT-3 and GPT-4. Musk has highlighted this change as evidence of mission drift, while Altman argues that the information-sharing landscape evolved as others started to commercialize AI.
The trial has also drawn attention to the financial stakes involved. If a court finds that OpenAI's board misled the public and donors about its nonprofit status, it could be forced to restructure or even divest some assets. Microsoft, which has invested over $13 billion, has a close partnership with OpenAI but is not a party to the suit. However, its deep involvement means any ruling could affect its access to OpenAI's technology and talent. Analysts believe that a win for Musk would likely force OpenAI to rotate its for-profit profits back into the nonprofit, altering the company's trajectory significantly.
Regardless of the verdict, the case highlights how quickly AI companies have had to adapt their structures. OpenAI’s journey from altruistic nonprofit to market-dominating startup is a cautionary tale about the tension between ideals and pragmatism. Eddy’s simple statement that “the documents tell the truth here” serves as a reminder that, in the midst of high emotions and conflicting narratives, the evidence remains the most reliable guide. The coming weeks will determine whether that evidence supports Musk or Altman, but the real impact will be on the future of AI governance.
Source: The Verge News