A California federal court has cleared the way for a jury trial to begin this week in Elon Musk’s legal battle against OpenAI and its chief executive, Sam Altman, yet the case that reaches jurors will be a shadow of the sweeping fraud indictment Musk first laid out. US District Judge Yvonne Gonzalez Rogers in Oakland has dismissed the core fraud and constructive fraud allegations, accepting Musk’s own late request to abandon those claims. Opening arguments are expected on Tuesday after jury selection on Monday, and the dispute now turns on two far narrower counts: abuse of charitable trust and unjust enrichment. Viewed from Washington, the ruling sets the stage for a rare courtroom scrutiny of the legal architecture that allowed a non-profit to incubate a commercial juggernaut now reportedly eyeing a trillion-dollar public listing.

Musk’s original 2024 complaint painted a narrative of betrayal, alleging that Altman, OpenAI president Greg Brockman, and Microsoft — the start-up’s most muscular backer — conned him and the public by gradually converting the organisation from its founding mission into a for-profit entity after his departure from the board in 2019. The initial suit contained 26 claims, but Musk moved to strip them down, reasoning that a leaner case would focus jurors’ minds on his central demand: that the company remain a force for broad human benefit rather than what he calls a “wealth machine”. Analysts in London note that such voluntary pruning, while rare in high-stakes litigation, reveals Musk’s confidence that the reputational pressure of a trial can serve his ends even if the original fraud narrative has crumbled.

The damages sought remain monumental. Musk is asking for $150 billion, a figure that, according to individuals familiar with the matter, would be directed to charitable purposes aligned with OpenAI’s founding ethos. The number is tethered to the staggering valuation attached to a potential initial public offering: Reuters has reported that OpenAI is preparing for an IPO that could value the firm at $1 trillion, a metric that, viewed from Moscow’s business press, underscores how the contest over artificial general intelligence is being fought with financial weapons that dwarf the GDP of many nations.

Curiously, the courtroom drama has already been preceded by a revealing piece of public theatre. A manufacturing entrepreneur, Jim Belosic, has paid $100,000 to unlock a paywalled podcast interview with Altman and Brockman conducted by journalist Ashlee Vance. The episode, part of Vance’s "Core Memory" series, had been hidden behind a paywall until Belosic, CEO of laser-manufacturing company SendCutSend, effectively bought it open for the world. The gesture, widely discussed in Silicon Valley, speaks to the almost insatiable public appetite for the unfiltered words of those building the most powerful AI systems, and to the lengths supporters will go to ensure those words circulate freely.

The trial will unfold against a backdrop of intensifying global debate over how to govern AI development. In stripping the case to its charitable-trust essence, the court has sharpened the question at the heart of the matter: whether the billions being generated by the commercialisation of non-profit research constitute a betrayal of the donors and the public who underwrote its early years. Should OpenAI’s for-profit transformation be found to have unjustly enriched Altman, Brockman, and Microsoft at the expense of a charitable purpose, the ruling could send tremors through a generation of hybrid entity structures. For now, as the Bay Area watches jury selection, the world waits to see whether a courtroom can succeed where regulators and ethicists are still fumbling — by defining the legal boundaries between altruism and empire in the age of thinking machines.