OpenAI co-founder recounts tension with Elon Musk during 2017 California court testimony

Greg Brockman, president and co-founder of OpenAI, described during the trial in Oakland, in Califórnia, a heated encounter with Elon Musk in 2017 that ended with an implied threat of physical assault. Brockman told the jury that the billionaire abruptly changed his behavior when his proposal to increase influence in the artificial intelligence company was rejected. The testimony is part of the second week of a month-long trial that examines a dispute between Musk and Sam Altman, co-founder and CEO of OpenAI.

Confronto under company control

Brockman rejected Musk’s attempt to gain more power in OpenAI during that 2017 meeting. Após refused, the tone of the conversation radically changed. “I actually thought he was going to hit me,” Brockman told the court, describing Musk’s reaction. The meeting ended shortly after with Musk announcing that it would stop the financing it had been offering to the company since its founding in 2015.

Mensagens’s text displayed by OpenAI’s lawyers shows that Musk had attempted to use incentives to persuade Brockman and another co-founder, Ilya Sutskever. An August 2017 message from Sutskever to Brockman asked: “Would a Model 3 make you accept extremely unfavorable conditions?” The strategies included what Brockman described as “curling up” to co-founders to gain greater control over the company’s operations.

Conhecimento preview on transition to for-profit

A crucial point of Brockman’s testimony is his assertion that Musk was aware of OpenAI’s plans to transform into a for-profit entity. The company was founded as a nonprofit, but later added a for-profit arm to raise billions in funding from investors. Há just over a year ago, the company decided to make the for-profit division its main focus. Brockman testified that Musk was aware of these intentions before leaving OpenAI, contradicting Musk’s defense claims.

The lawsuit revolves around Musk’s accusation against Altman of defrauding him of millions of dollars and having betrayed the original non-profit mission of the company responsible for ChatGPT. The court examines whether OpenAI’s transformation into a more traditional for-profit business violated original commitments or agreements established between the partners.

Detalhes about Shivon Zilis and conflict of interest

Brockman also revealed that Shivon Zilis, a previous OpenAI board member, is the mother of Mus’s four children, information he discovered through public reports. Quando asked how he learned this information, Brockman explained that Zilis informed him that he had had twins through in vitro fertilization and described the relationship as “completely platonic with Elon.” Later Questionamentos addressed Zilis’s continued role in OpenAI years after Musk’s departure.

Brockman stated that OpenAI trusted Zilis to “keep the conflict with Elon under control” during his years on the board. Essa trust was placed despite the personal relationship established between Zilis and Musk. Zilis left the board in March 2023, just as Musk was launching xAI, an artificial intelligence company developing a chatbot that competes directly with OpenAI’s ChatGPT.

Contexto from the biggest dispute

The legal battle reflects a dispute that has grown over the years since Musk left OpenAI, where he was an initial co-founder. Desde then transformed the company into one of the most valuable technology companies in the world, driven especially by the successful launch of ChatGPT. Musk accuses Altman of deceiving him while building a billion-dollar corporation while maintaining the appearance of a company focused on the common good.

The trial represents one of the biggest legal showdowns involving prominent figures in the technology and artificial intelligence industry. The court’s ruling in Oakland could have significant implications for how AI companies structure their operations and communicate missions to their founders and investors. The dispute also highlights questions about corporate governance and accountability when strategic visions differ significantly between co-founders.

Próximos trial steps

Espera Shivon Zilis is called to testify after Brockman ends his participation in the proceedings. Seu testimony may provide additional insights into Musk’s attempts to maintain influence within OpenAI and into the company’s internal dynamics during the critical period when decisions about its business structure were made.

The trial remains in its second week, with approximately two weeks of proceedings still scheduled. Evidence presented so far includes:

• Mensagens 2017 text between co-founders discussing incentive offers
• Documentos Insiders on Transitioning to For-Profit Model
• Testemunhos on Enterprise Transformation Prior Knowledge
• Registros communication between Musk and board members
• Evidências on conflicts of interest involving board members

OpenAI maintains that Altman acted within its rights in guiding the company through the structural transition, while Musk seeks to reverse the changes or obtain meaningful compensation for the rights it claims were violated. The court must determine whether there were binding agreements that were broken or whether OpenAI executives’ actions remained within established legal and contractual limits.