Musk v. Altman Trial Advances with Key Arguments Over OpenAI's Corporate Conversion

In a federal courtroom in Oakland, California, two of the most powerful men in technology are fighting over what a nonprofit promise made in 2015 means in a world where artificial intelligence companies are valued at hundreds of billions of dollars. The Musk v. Altman trial, which began on April 27 with jury selection and opening arguments, has already produced damning diary entries, tearful former executives, and an admission that Musk's own AI company trains on OpenAI's models [1][2].

The trial is expected to last five weeks, with a verdict from the nine-person advisory jury anticipated in late May or early June. But because this is an equitable claim — centered on breach of charitable trust — Judge Yvonne Gonzalez Rogers will ultimately decide the outcome herself [3].

At stake: up to $150 billion in potential damages, the removal of Sam Altman and Greg Brockman from their leadership roles, and the possible unwinding of OpenAI's for-profit conversion — a restructuring that transformed a research nonprofit into the most valuable private technology company in history [4].

From Zero to $852 Billion

When OpenAI was incorporated in December 2015, it held no meaningful assets beyond its founding team's ambition and roughly $1 billion in pledges from donors including Musk, Peter Thiel, Reid Hoffman, and others [5]. Its mission statement was explicit: "advancing digital intelligence in the way most likely to benefit humanity as a whole, unconstrained by a need to generate financial return" [6].

By April 2026, OpenAI had raised $122 billion in new funding at a valuation of $852 billion [7]. The nonprofit that once ran on donated compute time now operates a commercial empire anchored by ChatGPT and a sprawling partnership with Microsoft.

Source: TechCrunch, CNBC, OpenAI filings

Data as of Apr 30, 2026CSV

The restructuring that enabled this growth happened in stages. In 2019, OpenAI created a "capped-profit" subsidiary — OpenAI LP — controlled by the nonprofit board, with investors' returns limited to 100 times their initial investment [8]. In 2023, Microsoft invested $10 billion, becoming the company's largest financial backer [9]. By late 2025, OpenAI reached an agreement with the California and Delaware attorneys general to convert its for-profit arm into a public benefit corporation (PBC), with the newly created OpenAI Foundation retaining approximately 26% of the equity — a stake worth roughly $220 billion at current valuations [10].

Musk's expert witness, David Schizer, a former dean of Columbia Law School, testified that the Foundation should hold "a lot more" than $200 billion, arguing that the nonprofit gave up significant value to other stakeholders — particularly Microsoft — without conducting an adequate evaluation of what it received in return [11][12].

Source: OpenAI, CNBC

Data as of Apr 30, 2026CSV

The Three Hits

The trial's second week opened with testimony that Musk's legal team structured around three lines of attack against Altman's leadership.

Safety abandoned. Rosie Campbell, a former OpenAI safety researcher who worked at the company from 2021 to 2024, told jurors that OpenAI once had two dedicated long-term AI safety teams. One focused on aligning AI systems with human values; the other, where Campbell worked, prepared for the societal impact of superhuman intelligence. Both were eliminated. About half her team left the company rather than accept reassignment to other roles [13][14]. OpenAI also removed the word "safely" from its mission statement in a November 2025 IRS disclosure, changing its stated aim from building AI that "safely benefits humanity" to simply building AI that "benefits humanity" [15].

A culture of deception. Jurors heard edited portions of a videotaped deposition from Tasha McCauley, a former OpenAI board member. McCauley testified that Altman fostered a "culture of lying and culture of deceit" that had "trickled down" to other members of OpenAI's leadership [16]. She alleged that Altman falsely claimed the legal department had approved the launch of GPT-4 Turbo in India without safety board review. She described "repeated crisis events" stemming from Altman's behavior and said the board had "buckets of concerns" about his leadership, including resistance to oversight and lies reported by senior leaders [17].

Her testimony built on earlier depositions from fellow former board member Helen Toner, who said she first learned about ChatGPT's release from an employee rather than from Altman, and who referenced an email from former board member Ilya Sutskever documenting "dozens of pages of examples" of chaotic incidents caused by Altman's conduct [18].

Former CTO Mira Murati, testifying live, told the court that Altman said "one thing to one person and completely the opposite to another person." She said Altman pitted executives against each other and undermined her in her role [19]. Murati also testified that Altman misled her about safety clearances for a new model, falsely claiming OpenAI's legal team had determined it did not need deployment safety board review [20].

Nonprofit governance violated. Schizer testified that, assuming certain facts in evidence are proven, OpenAI failed to follow customary practices of nonprofit corporations in multiple areas — particularly in the relationship between the board and CEO. "The board and CEO need to be partnering, working together, to make sure the mission is being followed," Schizer said. "If the CEO is withholding that information, it's a big problem" [12]. He focused on the 2023 Microsoft investment, arguing the nonprofit gave up value without proper evaluation [11].

The Diary That Became Exhibit A

The most damaging evidence against OpenAI's leadership did not come from Musk's witnesses. It came from the personal journal of Greg Brockman, OpenAI's co-founder and former president.

In a November 2017 entry, Brockman wrote about being "warm to steal the nonprofit from [Musk] to convert to b corp without him" [21]. Another entry read: "Cannot say that we are committed to the nonprofit. Don't wanna say that we're committed. If three months later we're doing b-corp then it was a lie" [22].

Brockman's journal also included a self-aware reflection: "[Musk's] story will correctly be that we weren't honest with him in the end about still wanting to do for profit just without him" [21].

OpenAI's lawyers argued these entries are "staged for maximum misrepresentation," cherry-picked from hundreds of pages of unguarded personal reflections that include self-doubt and introspection [22]. Brockman, who took the stand during the trial's first week, acknowledged he never fulfilled a $100,000 donation pledge to the nonprofit — while now holding approximately $30 billion in for-profit equity [23].

During the same period, in 2017, Altman wrote in an email that he remained "enthusiastic about the non-profit structure" after Musk threatened to cut off funding [4].

The Competitor in the Courtroom

OpenAI's defense rests heavily on the argument that Musk's lawsuit is a commercial weapon, not a charitable crusade. Attorney William Savitt told the jury that Musk "didn't get his way" — that he had pushed for a for-profit structure himself, left when he couldn't exert full control, and filed suit only after founding xAI as a direct competitor in 2023 [24].

The conflict of interest argument has teeth. xAI, which builds the chatbot Grok, was sold to SpaceX for $250 billion and is expected to go public as part of SpaceX as early as June 2026 [25]. Tesla invested $2 billion in xAI's Series E round, and xAI purchased $430 million in Tesla Megapack battery systems in 2025 [25].

Musk acknowledged on the stand that xAI "is, at this point, technically a competitor, though much smaller than OpenAI" [24]. He also admitted that xAI trains its Grok model by "distilling" OpenAI's models — a practice he characterized as something "everybody does" [2].

Musk's legal team has tried to neutralize the conflict argument by requesting that any damages be paid not to Musk but to the OpenAI Foundation itself. In an April filing, his attorneys said "ill-gotten gains" should be returned to the charitable entity [26]. But the fact remains: if OpenAI's conversion is blocked or its operations are frozen by litigation, xAI benefits directly from a weakened competitor.

The Blue Cross Precedent

The legal fight over OpenAI's conversion is not without precedent, though no prior case has involved assets of this scale.

In the 1990s, Blue Cross of California's attempted conversion from nonprofit to for-profit status triggered a three-year legal battle. California regulators ultimately required the creation of two multibillion-dollar charitable foundations to preserve the value of charitable assets [27]. Blue Cross and Blue Shield of Missouri faced similar litigation in 1996 when it placed 80% of its assets in a for-profit subsidiary [27]. In Georgia, a citizen-group lawsuit forced Blue Cross Blue Shield to create a foundation holding 20% of the new corporation's stock [27].

The legal doctrine underlying these cases — "charitable trust" — requires that assets of a charitable nonprofit must remain dedicated to the charitable purposes for which they were established, even if the corporation is dissolved or reorganized [27]. Under the related "cy pres" doctrine, when original charitable purposes become impracticable, assets may be redirected to similar purposes [27].

Musk's surviving claims rest on this framework. The January 2026 order denying OpenAI's motion to dismiss found triable disputes of fact as to whether Musk attached enforceable terms to his approximately $38 million in seed donations — specifically, that OpenAI remain nonprofit and open-source [28].

Legal scholars have noted a critical distinction that public commentary often overlooks: a nonprofit's corporate purpose is not the same thing as a restricted charitable gift. The case turns on whether Musk can prove his donations carried enforceable charitable restrictions, whether OpenAI's restructuring breached those restrictions, and whether defendants retained benefits that should be restored to the charitable purpose [28].

What California Law Actually Requires

Under California nonprofit corporation law, charitable assets must be "permanently dedicated to a charitable purpose" [29]. The California Supreme Court has held that all assets held by a corporation organized for charitable purposes are "deemed held in trust for that declared purpose" [29].

The California Attorney General's office investigated OpenAI's conversion and, in October 2025, reached a memorandum of understanding that allowed the restructuring to proceed — provided the OpenAI Foundation retained its 26% equity stake and maintained governance independence [30]. Critics, including Public Citizen, argued this arrangement was insufficient. They pushed for the company to disband its nonprofit entirely, evaluate the full value of its assets, and donate an equivalent amount to independent charities — the same approach California required of Blue Cross in the 1990s [31][29].

The Columbia Law School Blue Sky Blog published an analysis arguing that hybrid models combining for-profit incentives with nonprofit oversight "suffer from structural incoherence, having the limitations of both organizational forms while securing neither's benefits" [32].

Employees, Investors, and the Microsoft Question

The conversion's outcome has direct consequences for thousands of people. OpenAI employs over 3,000 workers, many of whom hold equity in the for-profit subsidiary [33]. Investors who have poured hundreds of billions into the company — most recently $122 billion in April 2026 — hold stakes contingent on the corporate structure remaining intact [7].

Microsoft, OpenAI's largest shareholder at 27%, restructured its partnership in late April 2026. Under the new terms, Microsoft will no longer pay a revenue share to OpenAI, while OpenAI's revenue share payments to Microsoft continue through 2030, subject to a cap. Microsoft also no longer needs to determine its response if OpenAI declares it has achieved artificial general intelligence [34]. The revised deal is widely viewed as IPO-friendly, though analysts have shifted the realistic IPO window to mid-to-late 2027 [35].

If the court blocks or substantially delays the conversion, the consequences ripple outward. Employee equity could be frozen in legal limbo. Investor returns — structured around the for-profit entity — face uncertainty. OpenAI has argued in filings that disrupting the conversion would impair its ability to raise capital and compete in a market where rivals including Google DeepMind, Anthropic, and xAI are spending aggressively on compute and talent [33].

If Musk Wins

The most difficult question the trial raises is not legal but practical: what happens if Musk prevails?

If the conversion is unwound, OpenAI would revert to a structure that its own leadership has argued is inadequate for a company operating at its current scale. A nonprofit cannot easily attract the capital needed to fund the billions in infrastructure spending that frontier AI development demands. Competitors — none of whom face similar structural constraints — would advance while OpenAI sorts through governance questions in court.

If damages are awarded to the OpenAI Foundation, the charitable entity gains resources but faces the same structural question: how does a nonprofit with a $200+ billion endowment effectively oversee a technology company that requires commercial-speed decision-making?

If Altman and Brockman are removed, the leadership vacuum would arrive at a moment when the company is mid-IPO preparation, negotiating global regulatory frameworks, and racing to build next-generation AI systems.

Musk's framing — "you can't just steal a charity" — is morally intuitive. But the remedy he seeks could produce an outcome where the charitable mission is nominally preserved while the organization that built the technology either stalls or fragments, and competitors without any nonprofit obligation fill the gap [1].

What the Trial Is Really About

Beneath the personal drama between two billionaires, the Musk v. Altman trial tests a question that extends well beyond OpenAI: can the legal framework designed for hospital conversions and Blue Cross plans govern the most consequential technology companies of the 21st century?

The charitable trust doctrine was built for a world where nonprofit assets were buildings, endowments, and insurance pools. OpenAI's assets are trained neural networks, research talent, and partnerships with hyperscale cloud providers. Whether a doctrine designed for the conversion of community hospitals can meaningfully constrain a company valued at $852 billion is the question the trial will ultimately answer — not just for OpenAI, but for every nonprofit that incubates commercially valuable technology.

The jury will render its advisory verdict. Judge Gonzalez Rogers will decide the rest. And the AI industry, which has watched this trial with a mixture of fascination and anxiety, will learn whether the promises made in founding documents actually bind the companies that outgrow them.