On April 28, local time, Tesla CEO Elon Musk and OpenAI CEO Sam Altman appeared in the U.S. federal court at the same time to face off over OpenAI’s development over the past decade and its future direction. This was originally regarded as an important moment in the field of AI safety, but the court battle between the two gradually deviated from the important issue of "what should AI be oriented towards" and evolved into a battle driven by personal grievances and money.
Musk sued OpenAI, Altman and others on March 29, claiming that it violated the company's original mission of developing artificial intelligence for human welfare rather than profit.
OpenAI's lawyers told the jury that OpenAI never promised to remain a non-profit or make all of its code public. "The evidence will show that what Musk said did not happen."
Musk, the first person to testify in the case, tried to break through the discussion of OpenAI itself. He told the panel of nine jurors that OpenAI's transformation of the artificial intelligence laboratory into a for-profit company valued at $850 billion was equivalent to "stealing a charity." If he supports Altman, it "will give a green light to plunder all charities in the United States."
Musk has filed at least four different lawsuits against OpenAI in recent years, many of which have been withdrawn or dismissed. But this time, he named Altman, OpenAI President Greg Brockman, Microsoft and OpenAI itself as defendants, and many leaders in the AI industry are expected to testify.
"The tail wags the dog" and "the dog that doesn't bark"
During the court session that day, OpenAI’s lawyers complained that Musk issued an inflammatory “series of tweets” on April 27 and promoted a “malicious article,” referring to a recent New Yorker report questioning Altman’s character.
When the judge asked Musk, he said he was responding to OpenAI's previous post. The judge then asked both parties to "clear up" and Musk and Altman agreed to stop using social media during the trial.
Musk's lawyer, Steven Morrow, said that Musk has been worried about computers becoming smarter than humans since his college days. He has lobbied the government to formulate regulatory policies to deal with general artificial intelligence (AGI), including meeting with then-President Obama in 2015, but the government did not take action. "Elon felt he had to do something on his own."
Around the same time, Musk met Altman, a 30-year-old investor at the time, and they soon co-founded OpenAI, initially as a nonprofit. Google's rapid progress in AI has sparked concerns among its two co-founders, who want to build a competing research organization that focuses more on security.
"In my opinion, OpenAI exists because Larry Page (Google CEO) called me a 'speciesist' because I support humans." Musk said, "So what is the opposite of Google? An open source non-profit organization."
With OpenAI enjoying some success, Musk and Altman agreed to set up a "limited return" for-profit unit to raise huge sums of money for hiring and computing resources. The attorney likened it to a nonprofit museum getting some of its revenue through a store. "As long as the tail doesn't wag the dog, I'm not opposed to having a small for-profit division," Musk said in court.
"The tail wags the dog" is an English proverb. In the business field, it usually refers to a subsidiary department of a company that determines the strategic direction of the entire enterprise.
As Microsoft invests $10 billion in 2023 and OpenAI moves more and more of its intellectual property and employees to for-profit companies, Musk thinks the company has gone too far. “It’s like a museum store selling Picassos and locking them away from the public,” Morrow said.
Musk claimed that he was misled into donating approximately $38 million under the belief that OpenAI would remain a non-profit. He accused Altman and Brockman of improperly profiting by turning the startup into a for-profit enterprise.
OpenAI argued that it had “no record of these commitments.” "This is a dog that doesn't bark," his lawyer William Savitt told the court, suggesting that if there had been a promise there would have been evidence, but there wasn't.
"The reason why we are here is because Mr. Musk did not get the results he wanted at OpenAI...because he is now using his own company xAI to compete with OpenAI...just because he is a competitor, he will do everything possible to attack OpenAI." Savitt said that Musk "cannot afford to lose."
According to OpenAI, it was Musk himself who failed to fulfill his promise. He had pledged to invest up to $1 billion in the company, but only committed about $38 million over five years. In addition, although Musk claims to want to prevent superintelligence from falling into the hands of a single organization or individual, he has proposed that OpenAI be controlled by himself or Tesla.
“I would rather live in a world like Star Trek”
"AI could kill us all... a Terminator-style ending. I think we would rather live in a world like Star Trek than in a James Cameron movie," Musk said in court.
Although Musk has long warned about AI safety, his company xAI was once criticized by researchers from other laboratories for not paying enough attention to safety risk control. xAI's chatbot Grok has caused disturbing incidents, with thousands of people using it to "undress" images of women on social platform X.
Musk asked the court to "deprive Altman and Brockman of their management rights and the personal economic benefits they derived from OpenAI's illegal profit-making operations and transformation" and to revoke OpenAI's for-profit restructuring. The court is expected to pronounce its verdict in late May.
If OpenAI wins the lawsuit, the agency can continue to advance on its existing monetization path. If Musk wins the case, he will severely damage OpenAI because it will struggle to attract huge investments once it loses its for-profit division.
Recently, OpenAI has also faced financial difficulties. According to a report by The Times on the 28th, OpenAI has failed to achieve its new user and revenue targets in recent months. OpenAI Chief Financial Officer Sarah Fryer expressed concerns to the company's leadership about whether the company will be able to pay for future computing power contracts if its revenue does not grow fast enough.