Elon Musk took the witness stand in a federal courtroom in Oakland this week and did something no one in Silicon Valley has managed to do in three years of AI hype: he made Sam Altman look like the bad guy. Musk’s $130 billion lawsuit against OpenAI — the single largest damages claim in tech history — isn’t just a grudge match between two billionaires. It’s a trial that could legally define whether an AI company can be born as a charity, raise billions on that reputation, and then quietly pocket the proceeds.
And based on what came out in the first three days of testimony, OpenAI’s defence is in far worse shape than the company’s PR team has been letting on.
Musk’s Opening Shot Was a Founding Document OpenAI Wishes Didn’t Exist
Musk’s legal team entered into evidence the 2015 founding charter of OpenAI — the original document that declared the organisation would build “open source technology for the public benefit” and was “not organised for the private gain of any person.” That’s not a vague mission statement buried in a blog post. It’s a legal commitment made to donors, the IRS, and the public.
Then Musk testified about his own role in creating the organisation. “I came up with the idea, the name, recruited the key people, taught them everything I know, provided all the initial funding,” he told the nine-person jury empanelled by Judge Yvonne Gonzalez Rogers. Whether or not you believe every word of that — and there are reasons to be sceptical about Musk’s tendency to claim credit — the paper trail largely backs him up. Early emails, board minutes, and financial records show Musk as the primary early funder who recruited Ilya Sutskever and helped shape the nonprofit’s original direction.
The point isn’t whether Musk is a saint. The point is that OpenAI accepted his money — and everyone else’s — under a specific legal framework. And then it changed the rules.
The Microsoft Deal Is Where This Gets Ugly
The most revealing moment of Musk’s testimony came when he described his reaction to learning about Microsoft’s $10 billion investment in OpenAI back in late 2022. “I reacted quite negatively,” Musk told the court, “because at a $10 billion scale, there’s no way Microsoft is just giving that as a donation or any kind of charitable way.”
That’s the core of the case in one sentence. Microsoft didn’t write a $10 billion cheque because it cared about open-source AI for humanity. It wrote that cheque because it expected a return — and OpenAI’s leadership structured the deal to make sure it got one. The nonprofit shell was kept around as a governance fig leaf while a for-profit subsidiary was created to hoover up the actual value.
Musk’s lawyer, Marc Toberoff, put it more bluntly in his opening statement: OpenAI’s leadership “stole a charity.” That’s a dramatic framing, but it maps cleanly onto what happened. OpenAI built its reputation and recruited its talent under a nonprofit banner. It then created a capped-profit entity, attracted tens of billions in investment, ballooned to a $500 billion-plus valuation after its 2025 tender offer, and is now attempting to convert fully into a for-profit corporation.
If that sounds like having your cake and eating it too, that’s because Musk used those exact words on the stand.
What Musk Wants — And Why It Matters Beyond the Money
Musk isn’t just asking for $130 billion in damages. He wants the court to force OpenAI back to a nonprofit structure and to remove Sam Altman and Greg Brockman from the board. That second demand is the one that should make every AI executive pay attention.
If Judge Gonzalez Rogers rules that OpenAI’s leadership breached its fiduciary duties to the original nonprofit — and legal analysts watching the trial say at least one count is likely to stick — it sets a precedent that goes far beyond this case. It would mean that AI organisations can’t use nonprofit status as a launch vehicle, build up enormous value on the back of that credibility, and then convert to a for-profit structure without accountability to the original donors and charter.
That has implications for every AI lab that has flirted with nonprofit-to-commercial transitions. Anthropic, which spun out of OpenAI and structured itself as a public benefit corporation, would be watching this trial with particular interest. So would the growing number of AI startups that have used nonprofit research arms to attract grants and talent before routing the commercial value elsewhere.
The Elephant in the Courtroom: xAI
Here’s the part nobody in Musk’s camp wants to talk about, and that OpenAI’s lawyers will absolutely hammer in cross-examination: Musk runs xAI, a direct competitor to OpenAI. He has a massive financial incentive to hobble OpenAI’s commercial operations, and a $130 billion damages award would be the most effective way to do it short of an act of Congress.
OpenAI’s defence team is expected to argue that Musk’s lawsuit is fundamentally a competitive move dressed up in idealistic language. And they’re not entirely wrong. Musk left OpenAI’s board in 2018, reportedly after a failed bid to take personal control of the organisation. The lawsuit landed only after xAI was up and running and ChatGPT had become the fastest-growing consumer product in history.
But here’s the thing about motivations: they don’t change the facts. Even if Musk is suing because he wants to kneecap a rival, the founding charter still says what it says. The nonprofit commitments were still made. The for-profit conversion still happened. A jury doesn’t have to like Musk to find that OpenAI broke its own rules.
The Verdict Could Land by Mid-May — And the Losing Side Will Appeal for Years
Judge Gonzalez Rogers is expected to deliver her ruling by mid-May 2026, after weighing the advisory jury’s findings. But regardless of which way it goes, the loser will appeal to the Ninth Circuit, where the case would likely be heard in late 2026 or early 2027. This legal battle has years left to run.
What matters right now isn’t the final verdict. It’s what’s already been entered into the public record. The founding charter. The emails. The financial structure that took a nonprofit’s goodwill and funnelled it into a half-trillion-dollar commercial enterprise. That evidence doesn’t go away regardless of the outcome.
Microsoft, which has poured tens of billions into OpenAI and built its entire Copilot strategy around the partnership, has stayed conspicuously quiet during the trial. But if a damages award in the tens of billions lands — even well short of the $130 billion headline number — it would throw OpenAI’s capitalisation structure into chaos and potentially force a renegotiation of Microsoft’s commercial terms.
The Real Question This Trial Answers
Strip away the personalities and the billions and the courtroom drama, and this trial is really about one question: can you build a company worth half a trillion dollars on the back of a charity and then keep the money?
Sam Altman’s answer is yes, as long as you restructure carefully and argue that commercialisation serves the original mission. Elon Musk’s answer is no, and he’s willing to spend years in court to prove it. The nine jurors in Oakland — none of whom work in tech, none of whom have a financial stake in AI — will deliver the first answer that actually carries legal weight.
Whatever they decide, the era of AI companies using nonprofit status as a convenient costume just ended. The only question now is how much it costs.
