Sam Altman takes the stand as Musk's OpenAI lawsuit reaches its pivotal week

The case, being heard before US District Judge Yvonne Gonzalez Rogers in federal court in Oakland, California, centres on whether Altman and OpenAI president Greg Brockman breached a charitable trust when they restructured OpenAI away from its nonprofit origins.

Musk, who co-founded OpenAI in 2015 and donated approximately $38 million to the project, is seeking $150 billion in damages and asking the court to revert the company to nonprofit status with both men removed from their roles.

Altman’s appearance follows three days of testimony from Musk himself in the trial’s first week, and a day on the stand on Monday from Microsoft chief executive Satya Nadella, the first public testimony from the head of OpenAI’s largest backer. Nadella told the court that Musk never directly raised concerns with him about whether Microsoft’s investments in OpenAI violated any special terms, and described the OpenAI board’s handling of Altman’s brief ousting as CEO in 2023 as “amateur city.”

Microsoft, which has invested more than $13 billion in OpenAI across three tranches since 2019, is named as a co-defendant on a charge of aiding and abetting the alleged breach of charitable trust.

Musk’s central argument is that OpenAI’s 2015 founding charter, which declared it would pursue “open-source technology for the public benefit” and was “not organised for the private gain of any person”, established a binding charitable trust, and that Altman and Brockman violated it by allowing the for-profit subsidiary to become the dominant entity.

He told the jury: “If we make it OK to loot a charity, the entire foundation of charitable giving in America will be destroyed.” He identified Microsoft’s $10 billion investment in 2022 as the moment he concluded the mission had been abandoned, describing it as self-evidently too large to constitute a donation.

OpenAI’s defence has been equally direct. Lead counsel William Savitt told jurors that Musk himself sought to lead OpenAI as a commercial entity and sued only after failing to secure what Savitt called “the keys to the kingdom.”

Savitt pointed to Musk’s subsequent founding of xAI, now merged with SpaceX, as evidence of motive. During cross-examination Musk acknowledged that xAI had partly used OpenAI’s technology to train its own models, describing the practice as industry standard.

The most damaging material for OpenAI has come from Brockman’s own 2017 journal entries, introduced into evidence by Musk’s lawyers, in which he wrote about being “warm to steal the nonprofit from [Musk] to convert to b corp without him” and acknowledged that Musk’s account of events would “correctly be that we weren’t honest with him.” Brockman also admitted to holding an undisclosed stake in AI chip startup Cerebras while OpenAI was considering acquiring the company.

The jury’s verdict is advisory; Judge Rogers makes the final determination. Most legal analysts consider Musk’s case weak, and OpenAI, now valued at more than $850 billion by private investors, has called his allegations baseless. But for the digital infrastructure and connectivity sector, the trial’s significance extends beyond the courtroom.

The Microsoft-OpenAI partnership underpins a substantial share of AI cloud capacity across Europe and globally.

A finding against Microsoft, even a partial one, would introduce structural uncertainty into a relationship on which billions of dollars of enterprise connectivity and data centre investment is currently predicated.

xAI data centre ‘Colossus’ found to be generating extra electricity illegally