The courtroom drama between Elon Musk and Sam Altman has entered a phase that is easy to overlook if you’ve been following the trial mainly for the headline-grabbing testimony. But this is where the case can start to matter in a very practical way: not just who said what, or which version of OpenAI’s origin story the jury believes, but what the court could order if Musk’s claims succeed.
After closing arguments wrapped up last week, the parties are back in court on Monday, May 18, for a hearing focused on possible remedies. In other words, the legal question shifts from “what happened?” to “what should change, if anything?” That distinction sounds procedural, yet it’s often the moment when governance disputes stop being abstract and become concrete—potentially affecting how OpenAI is structured, how leadership decisions are made, and how the organization’s public-interest commitments are enforced.
To understand why this remedies hearing is being watched so closely, it helps to revisit the core conflict. Musk, a cofounder of OpenAI who now leads rival xAI, argues that OpenAI abandoned its founding mission: developing AI to benefit humanity rather than pursuing profit as the primary driver. His lawsuit contends that Altman and Greg Brockman—both central figures in OpenAI’s modern era—steered the organization away from its original nonprofit/public-benefit orientation. Musk’s requested remedies go beyond symbolic statements. He has asked for changes that would remove Altman and Brockman from their roles and for OpenAI to stop operating as a public benefit corporation. He has also demanded damages that, if awarded, could reach up to $150 billion for the nonprofit entity.
OpenAI’s response is equally pointed, though it frames the dispute differently. OpenAI says the lawsuit is baseless and characterizes it as a competitive attempt to derail a rival. In OpenAI’s telling, Musk’s motivations are not primarily about mission fidelity; they’re about undermining OpenAI while boosting his own companies—an argument that has gained traction in the public narrative because xAI’s Grok has been positioned as a competitor to ChatGPT. OpenAI’s position also disputes the underlying premise that the organization’s evolution represents a betrayal of its mission rather than a pragmatic response to the realities of building frontier models at scale.
The trial itself has already featured testimony from both sides’ key witnesses, including Musk and prominent OpenAI leadership figures. According to the reporting summarized in the materials you provided, Musk, Jared Birchall (Musk’s financial manager and Neuralink CEO), and OpenAI cofounder Greg Brockman have testified before the jury. Shivon Zilis—described as a former OpenAI board member who shares multiple children with Musk—also took the stand, and the courtroom reviewed video deposition material from former OpenAI CTO Mira Murati. For the third week of testimony, Microsoft CEO Satya Nadella appeared first, followed by Ilya Sutskever, a former OpenAI cofounder and chief scientist. Altman then took the stand on Tuesday to refute Musk’s portrayal of him, including the argument that Musk’s accusations are rooted in personal animus rather than facts.
That sequence matters because it shows how the trial has been trying to answer two different questions at once. One question is factual: what did people know, when did they know it, and what decisions were actually made? The other question is interpretive: what do those decisions mean for the organization’s mission and governance obligations? Remedies hearings tend to amplify the second question, because courts must translate contested narratives into enforceable outcomes.
In the remedies phase, the court will be asked to consider what relief is legally available and practically feasible. That’s where the stakes become less about courtroom theatrics and more about institutional design. If Musk’s requested changes were granted, the implications would extend beyond the individuals named in the complaint. Removing specific leaders and altering corporate status would reshape how OpenAI’s governance works, potentially affecting board composition, decision-making authority, and the mechanisms by which mission-related commitments are monitored.
But even if the court is sympathetic to Musk’s framing, remedies are constrained by law. Courts generally cannot rewrite an organization’s entire structure based solely on a jury’s view of intent; they must tie any ordered relief to the legal claims that were proven. That means the remedies hearing is likely to focus on the fit between the alleged wrongdoing and the requested cure. It’s one thing to argue that OpenAI drifted from its mission; it’s another to persuade a judge that the specific remedy sought—leadership removal, a shift away from public benefit corporation status, and massive damages—is the appropriate and proportionate response.
This is why the remedies hearing is being treated as a pivotal moment. It’s not merely a continuation of the trial; it’s a pivot toward the question of whether the court can meaningfully enforce mission governance through structural changes. In disputes involving nonprofits, public-interest commitments, and hybrid corporate forms, remedies often become a proxy for a broader debate: can mission be compelled through legal architecture, or does mission enforcement inevitably collide with the realities of funding, competition, and operational complexity?
OpenAI’s defense suggests it believes the latter. By calling the lawsuit a competitive derailment attempt, OpenAI is implicitly arguing that the court should not treat mission drift allegations as a basis for sweeping governance intervention. OpenAI’s narrative also implies that the organization’s evolution—especially its relationship with major investors and partners—was not a betrayal but a necessity. In that framing, the remedies Musk seeks would be disruptive not only to OpenAI’s leadership but to its ability to operate effectively in a market where compute, talent, and distribution are tightly coupled to capital.
The presence of Microsoft executives and testimony about the Microsoft relationship underscores that point. The trial materials you provided indicate that Microsoft CEO Satya Nadella testified, and that the courtroom also heard evidence and arguments related to the OpenAI-Microsoft partnership. That matters because remedies that alter OpenAI’s governance could indirectly affect how OpenAI interacts with partners. Even if the court orders changes only within OpenAI’s internal structure, the practical effect could ripple outward into contracts, investment expectations, and strategic alignment.
There’s also a deeper governance issue embedded in the remedies question: what does it mean for an AI lab to be accountable to a mission when the lab is simultaneously dependent on commercial-scale resources? Frontier AI development is expensive, and the trial record described in your inputs includes discussion of fundraising and investment at enormous scale. Altman’s testimony, as summarized, included claims about OpenAI raising approximately $175 billion in investment. Whether or not every detail of that testimony is accepted by the jury, the existence of such scale is hard to ignore in any remedies analysis. A court considering structural changes must grapple with whether those changes would impair the organization’s capacity to secure funding and continue research.
That’s where the remedies hearing becomes more than a legal exercise. It becomes a referendum on whether mission-driven governance can survive contact with the economics of frontier model development. If the court orders changes that weaken OpenAI’s ability to attract capital or maintain operational continuity, the remedy could be self-defeating. Conversely, if the court refuses to order meaningful relief, critics of OpenAI’s evolution may argue that mission commitments are effectively unenforceable once an organization reaches a certain scale.
The trial’s witness list also hints at how the court may evaluate credibility and intent—factors that often influence remedies. The materials you provided include testimony from people tied to OpenAI’s internal decision-making and board-level concerns. Shivon Zilis’s testimony, for example, is described as involving issues around board notification and internal trust. Mira Murati’s deposition is described as addressing problems with Altman that persisted after his return to the company, and as characterizing OpenAI as being at catastrophic risk of falling apart when Altman was fired. These details matter because remedies are often shaped by the court’s view of whether the alleged conduct was isolated mismanagement or part of a broader pattern of governance failure.
If the court concludes that the proven issues were narrow, remedies may be limited. If the court concludes that the proven issues reflect a systemic governance breach tied to mission commitments, remedies could be broader. Either way, the remedies hearing is where the legal system tries to draw a line between “people disagree about strategy” and “the organization violated duties in a way that warrants structural correction.”
Another factor likely to come into play is the requested damages. Musk’s demand for up to $150 billion in damages is extraordinary in both scale and implication. Even if a plaintiff can theoretically argue for large damages, courts often scrutinize causation and quantification: what exactly was the harm, how directly was it caused by the defendant’s actions, and how should it be measured? In mission-governance cases, damages can become especially contentious because the harm is not always a straightforward financial loss. It can be framed as loss of mission integrity, loss of opportunity, or diversion of resources. Translating those concepts into a dollar figure is difficult, and remedies hearings frequently reveal how much of a damages claim is aspirational versus legally grounded.
OpenAI’s counter-narrative—that the lawsuit is a baseless bid to derail a competitor—could also influence how the court views damages. If the court believes the claims are not supported by the evidence in a way that establishes legal liability, damages may be reduced or denied entirely. But even if liability is established, the court may still limit damages to what is provable and proportionate.
It’s also worth noting that remedies hearings can be where procedural and strategic considerations surface. Parties may argue about what relief is even available under the relevant legal framework. They may also argue about timing—whether certain remedies would be inequitable given the organization’s current state, ongoing operations, and reliance interests. In disputes involving large institutions, courts often consider the disruption cost of remedies. That doesn’t mean disruption automatically wins; it means the court will weigh whether the remedy is workable and consistent with the organization’s continued functioning.
For readers, the temptation is to treat this as a simple binary: either Musk wins and OpenAI is forced to change