'Your Questions Are Designed to Trick Me': Inside Musk's Combative Testimony in the $150 Billion OpenAI Trial

On April 29, 2026, Elon Musk looked across the courtroom at William Savitt, the attorney representing OpenAI and Sam Altman, and said flatly: "Your questions are not simple. They're designed to trick me, essentially" [1]. The exchange capped a second day of testimony that veered between grandiose claims about the future of charitable giving in America and pointed disputes over how much money Musk actually gave to the company he helped create.

The trial — Musk v. Altman — opened on April 27 in the U.S. District Court for the Northern District of California in Oakland before Judge Yvonne Gonzalez Rogers, the same judge who presided over the Epic Games v. Apple case [2]. A nine-person jury will determine whether OpenAI breached its founding promise to serve humanity as a nonprofit, or whether Musk is a spurned co-founder using the courts to hobble a competitor [3].

The Accusation of 'Trickstery'

The flashpoint came during cross-examination when Savitt pressed Musk on his claim, made in an earlier deposition, that he had donated $100 million to OpenAI. Savitt pointed out that Musk's actual contributions totaled approximately $38 million — far less than the amount Musk had previously stated under oath [1]. Musk bristled, accusing Savitt of being "misleading" and calling his questions a form of "trickstery" [4].

The judge intervened, asking both sides to "calm down" after the exchange grew heated enough that Musk asked Savitt: "Can I answer this question without you interrupting my answer?" [5]. On at least one occasion, Musk acknowledged that some of his prior deposition statements were imprecise, telling the courtroom, "I literally was a fool" for putting money into OpenAI without more formal protections [6].

Savitt's cross-examination was designed to undercut a central pillar of Musk's case: that he donated to OpenAI on the explicit understanding it would remain a nonprofit forever. Savitt pointed to emails and meeting notes from 2017 in which Musk appeared to endorse the idea that OpenAI needed a for-profit component to compete with Google [1]. Musk responded that he was fine with a for-profit arm "as long as it was a subsidiary of the nonprofit" and the profits were capped. "What you can't have is the for-profit become the main event, and that's what we have here," he told the jury [7].

What the Lawsuit Actually Claims

The case has been narrowed to two causes of action: breach of charitable trust and unjust enrichment [3]. Musk's legal team argues that his $38 million in donations created a charitable trust obligating OpenAI to remain a nonprofit, open-source research organization dedicated to the safe development of artificial intelligence [8]. They contend that OpenAI's subsequent creation of a for-profit subsidiary, and its eventual restructuring into a for-profit public benefit corporation, constituted an unauthorized diversion of charitable assets [9].

The damages figure has shifted over the course of the litigation. Early filings sought up to $134 billion from OpenAI and its lead investor, Microsoft [10]. At trial, Musk's attorneys have framed the demand as approximately $150 billion in "wrongful gains" — but Musk has publicly pledged to direct any award to OpenAI's nonprofit foundation rather than keep it himself [11]. He is also asking the court to remove Altman and OpenAI President Greg Brockman from their leadership positions and to force the company to revert to nonprofit governance [10].

The Documentary Record

Hundreds of pages of emails, text messages, and internal documents have been entered into evidence, providing an unusually detailed look at the power dynamics that shaped OpenAI from its earliest days [12].

Among the most damaging exhibits for Musk: a 2017 email in which he directed the head of his family office, Jared Birchall, to register a for-profit public benefit corporation in OpenAI's name. Musk testified he did so "in case it was needed," but in the same email thread he stated he needed to have "control" of the proposed entity [7]. In another exchange, Musk told OpenAI's cofounders he would "no longer fund OpenAI until you make a firm commitment to stay nonprofit" — language his own attorneys cited as evidence that the nonprofit commitment was a condition of his giving [12].

For OpenAI's defense, the emails cut both ways. Savitt highlighted communications showing Musk was aware of internal discussions about a for-profit structure well before his 2018 departure from the board. More critically, Musk sought not merely a voice in governance but outright control: he outlined a plan where he would "unequivocally have initial control of the company," though he said the arrangement would be temporary [13].

Musk also made statements that strained credulity on the stand. He claimed during testimony that he had read the "very first part" of an early term sheet — a detail absent from any of his depositions [5]. When asked whether he had ever called an OpenAI employee a "jackass," Musk conceded "it's possible" this had happened more than once, then insisted he does not lose his temper or yell — an assertion contradicted by Walter Isaacson's 2023 biography [5].

The Money: $38 Million vs. $852 Billion

The financial gap between Musk's contributions and OpenAI's current valuation is central to both sides' arguments. Musk donated approximately $38 million to OpenAI before his 2018 departure, though he had initially pledged roughly $1 billion over a period of years [3]. His donations carried no formal governance rights — a fact OpenAI has emphasized repeatedly [14].

Source: Bloomberg, TechCrunch

Data as of Apr 30, 2026CSV

Since Musk left the board, OpenAI's valuation has risen from essentially zero as a nonprofit research lab to $852 billion following a $122 billion funding round completed in March 2026 — making it the most valuable private technology company in history [15]. The trajectory underscores the scale of what Musk argues was "looted" from the nonprofit: in his telling, insiders enriched themselves by converting charitable assets into private equity [16].

Source: CNBC, Bloomberg

Data as of Apr 30, 2026CSV

OpenAI's defense flips this framing. The company argues that the for-profit restructuring was the engine of its growth, that Musk was aware of and supported the transition, and that he only filed suit in 2024 after failing to take over as CEO and launching his own competing AI venture, xAI [14]. xAI has itself reached a valuation of $250 billion as a subsidiary of X.AI Holdings Corp, following a merger with X (formerly Twitter) that valued the combined entity at $1.25 trillion [17].

Standing, Statute of Limitations, and Legal Hurdles

Legal scholars have raised serious questions about whether Musk can bring a charitable trust claim at all. Jill Horwitz, a nonprofit law scholar at Northwestern University, has called Musk's standing "pretty puzzling," noting that actions to enforce charitable purposes are typically brought by state attorneys general, not individual donors [18]. The California Attorney General's office has been monitoring OpenAI's conversion but has not joined Musk's suit.

Microsoft, which invested $13 billion in OpenAI and is also a defendant, has raised a statute-of-limitations defense. Microsoft's attorney pointed to a September 2020 post on X in which Musk wrote that "OpenAI is essentially captured by Microsoft" — evidence, the defense argues, that Musk knew about the Microsoft partnership years before filing suit and waited too long to act [18].

Judge Gonzalez Rogers has signaled skepticism about certain remedies. She noted during pretrial proceedings that it would be "unprecedented" for a court in a private breach-of-charitable-trust suit to order the kind of structural overhaul Musk seeks — namely, reversing a corporate conversion and removing officers [18].

What Legal Experts Say About the Case's Weaknesses

The breach-of-charitable-trust claim faces a high threshold. To prevail, Musk's team must demonstrate that his donations were made with explicit, legally enforceable conditions — conditions that OpenAI's leadership subsequently violated. Charitable trust law generally requires more than a donor's subjective intent; it typically demands clear written terms establishing a restricted gift [18].

The unjust enrichment claim is somewhat more conventional but carries its own difficulties. Musk must show that OpenAI, Altman, and Brockman were unjustly enriched by his donations — a claim complicated by the fact that the $38 million represented a small fraction of OpenAI's total funding and that Musk received no equity or contractual governance rights in return [3].

Trial attorneys who follow nonprofit litigation say cases like this rarely succeed when brought by individual donors rather than regulators. The typical remedy in a charitable trust dispute is a court order requiring the organization to return to its stated mission, not a multibillion-dollar damages award to the donor or the donor's preferred beneficiary [18].

The Competitive Dimension

OpenAI's lawyers have consistently argued that Musk's lawsuit is driven by competitive interest, not principle. Musk founded xAI in July 2023 and launched its chatbot, Grok, as a direct competitor to OpenAI's ChatGPT [14]. The timing of the initial lawsuit in 2024, shortly after xAI's founding, has fed the narrative that the litigation serves as a market weapon.

Musk has rejected this characterization, testifying that he would benefit more from OpenAI failing than from suing it: "If I wanted to hurt OpenAI, I would just compete with them, which I'm already doing" [7]. He has framed the case in broader terms, telling the jury: "If we make it okay to loot a charity, the entire foundation of charitable giving in America will be destroyed" [16].

Financial analysts have noted that a ruling in Musk's favor could indirectly benefit xAI by imposing structural constraints on its largest competitor. If the court ordered OpenAI to unwind its for-profit conversion, the disruption could redirect talent, capital, and partnerships toward competitors including xAI and Anthropic, which already operates as a public benefit corporation without the baggage of a nonprofit-to-for-profit transition [19].

Implications for AI Governance

The trial's significance extends beyond the two parties. Several AI companies have adopted hybrid governance structures — capped-profit subsidiaries, public benefit corporations, or nonprofit-controlled for-profit arms — to balance the need for massive capital investment against stated commitments to safety and public benefit [19].

If Musk prevails on the charitable trust claim, the ruling could expose any nonprofit-adjacent AI organization that has taken restricted donations and later restructured to similar legal challenge. Anthropic, which was founded by former OpenAI executives and operates as a public benefit corporation, has attracted roughly $30 billion in funding. Other hybrid entities in the AI space collectively manage tens of billions more [19].

Musk's lawyers have argued that the precedent matters for all of American philanthropy, not just AI. But OpenAI's defense counters that a ruling for Musk would chill innovation by making it legally perilous for any nonprofit to adapt its structure as circumstances change — a flexibility that nonprofit law has traditionally protected [14].

What Comes Next

Musk was expected to take the stand for a third consecutive day on April 30, with OpenAI's attorneys finishing cross-examination before Musk's lawyers conduct redirect [7]. The trial is expected to run roughly two to three weeks, with jury deliberations on liability anticipated by mid-May [3].

The nine jurors seated on April 27 will ultimately decide whether Musk's $38 million created an enforceable obligation or whether, as OpenAI contends, there was "never a commitment to remain a nonprofit" [14]. Whatever the verdict, the trial has already produced a remarkable public record: the private emails, power struggles, and broken relationships behind the most consequential AI company in the world.