Sam Altman says Elon Musk wanted 90 percent of OpenAI in high-stakes trial

In a United States court, OpenAI main enforcement Sam Altman has rejected claims from chap tech mogul Elon Musk that helium betrayed the artificial quality company’s archetypal vision.

Tuesday marked the commencement of Altman’s grounds successful a contentious proceedings unfolding successful Oakland, California, betwixt immoderate of tech’s richest and astir almighty titans.

Recommended Stories

• list 1 of 3Pentagon announces woody with 7 AI companies for classified systems
• list 2 of 3Microsoft, Google, xAI springiness US entree to AI models for information testing
• list 3 of 3SpaceX backs Anthropic with information centre woody amidst Musk’s OpenAI lawsuit

Musk, the wealthiest antheral successful the world, has sued Altman and OpenAI president Greg Brockman connected the ground that they “stole a charity” by shifting its purpose.

He alleged that OpenAI’s person persuaded him to put $38bn, based connected a extremity of improving humanity, lone to spot the institution pivot to a for-profit task successful 2019.

On the witnesser basal connected Tuesday, Altman alternatively framed Musk arsenic a rival obsessed with exercising power implicit OpenAI.

“It does not acceptable with my conception of the words ‘stealing a charity’ to look astatine what has really happened here,” Altman told the court.

The 2 men person agelong had an acrimonious relationship, driven successful portion by differing views astir artificial intelligence.

Musk — a self-described escaped code “absolutist” — presently runs his ain AI chatbot, Grok, which has been accused of perpetuating right-wing conspiracy theories and violative materials.

He is seeking $150bn successful damages from OpenAI and Microsoft, 1 of its main investors.

Altman’s grounds comes much than 2 weeks into the trial, which has seen him and Musk square off against each other.

In his testimony, Altman argued that Musk knew of the plans to make OpenAI into a for-profit endeavor erstwhile helium invested, and helium asserted that Musk adjacent petitioned to person a bulk involvement successful the company.

“An aboriginal fig that Mr Musk threw retired was that helium should person 90 percent of the equity to start,” Altman told the jury. “It past softened, but it ever was a majority.”

The result of the proceedings could find the aboriginal of OpenAI, its leadership, and products similar ChatGPT. As portion of his lawsuit, Musk is pushing for the removal of Altman and Brockman.

The proceedings comes arsenic OpenAI prepares for a imaginable archetypal nationalist offering that could spot it valued astatine $1 trillion, a historically ample sum.

During earlier testimony, Musk portrayed Altman arsenic a liar who could not beryllium trusted with the improvement of the technology.

“If you person idiosyncratic who is not trustworthy successful complaint of AI, I deliberation that’s a precise large information for the full world,” Musk said.

Musk’s lawyer, Steven Molo, besides sought to undermine Altman’s reliability during questioning connected Tuesday.

“Have you misled radical erstwhile you bash business?” Molo asked Altman.

“I bash not deliberation so,” Altman replied.

Altman, meanwhile, sought to formed uncertainty connected Musk’s leadership; Musk yet near OpenAI’s committee successful 2018 to prosecute his ain AI development.

“I don’t deliberation Mr Musk understood however to tally a bully probe lab,” Altman said. “He had demotivated immoderate of our astir cardinal researchers.”

The US public, for its part, has been largely unconvinced by high-minded rhetoric astir the transformative imaginable of AI.

A March 2026 canvass by the Pew Research Center suggested that a bulk of respondents successful the US judge AI volition worsen, alternatively than improve, the quality to deliberation creatively, signifier meaningful relationships, marque hard decisions, and lick problems.

Just 10 percent of respondents said they were much excited than acrophobic astir the accrued usage of AI successful regular life.

But the manufacture has been speedy to construe its important economical powerfulness into governmental power arsenic lawmakers see however champion to modulate the technology.

The usage of AI has emerged arsenic an election-season contented arsenic the US midterms attack successful November, and the medication of President Donald Trump has projected a “national argumentation framework” for the exertion to debar a patchwork of authorities regulations.

The AI manufacture has go a operator of eye-watering concern successful caller years, with the United Nations estimating that the planetary marketplace could beryllium worthy $4.8 trillion by 2033.