Late this February, Elon Musk sued OpenAI CEO Sam Altman, under allegations of the company deviating from its founding charter, amounting to a breach of contract. His complaint, filed in California state court suggests that ChatGPT diverged from its original nonprofit mission through its partnership with Microsoft, who came on board with $13 billion. According to Musk, OpenAI is also keeping the latest version of its code for its latest generative AI products a secret from the public.
Elon Musk was a founding member of OpenAI in 2015, before parting over disputes with the direction the organisation was taking. He has started his own AI company, x.AI, since. When it was first created, OpenAI was intended as a check on the market forces that threatened to expose humanity to a dangerous side of generative AI. This was ensured by having a board of overseers that included independent members to review the company’s products and ensure that their codes were made public. This was countered by Altman, Brockman, and Ilya Sutskever by creating OpenAI LP in 2019—a profit entity that existed within the larger company structure. This manoeuvre resulted in OpenAI going from a not-for-profit to a $90 billion valuation in a span of a few years.
Musk’s complaint with the court also cites another industry sleight of hand that took place when Sam Altman was ousted from his leadership position in the country, over concerns about his decisions that ignored the threats of artificial intelligence in the drive for profit. Microsoft’s intervention on Altman’s behalf resulted in his reinstatement, a move that industry analysts widely acknowledge as a victory for those who seek to commercialise AI. Musk’s complaint reads:
“The public is still in the dark regarding what exactly the Board’s ‘deliberative review process’ revealed that resulted in the initial firing of Mr. Altman… However, one thing is clear to Mr. Musk and the public at large: Open AI has abandoned its ‘irrevocable’ non-profit mission in the pursuit of profit.”
Musk’s complaint with the legal system also makes special mention of the role played by Microsoft in enabling this apparent deterioration of OpenAI’s initial motivations. While the tech giant is not named as a defendant in the initial lawsuit, it is mentioned close to 70 times in the documents themselves. The lawsuit names and criticises Microsoft and its influence over OpenAI, particularly in light of its comparative economic position:
“OpenAI, Inc. has been transformed into a closed-source de facto subsidiary of the largest technology company in the world: Microsoft… Under its new Board, it is not just developing, but actually refining an AGI (Artificial General Intelligence) to maximise profits for Microsoft.”
Musk’s quarrel with OpenAI is not limited to the legal sphere, given the SpaceX founder’s penchant for taking things public. In an X (formerly Twitter) post, Musk challenged OpenAI to change its company name to ClosedAI, promising to drop his lawsuit if they did. “OpenAI needs to stop living a lie”, said the voracious X poster, alongside a post of a photo of Sam Altman with an edited guest ID card that introduced the company as ‘ClosedAI’. According to legal experts, Musk’s lawsuit is as valid as his X posts. This is because there is no actual contract or other legal written document that OpenAI is in breach of, especially not one with the signatures of all the parties involved. Instead, they argue that this high-profile lawsuit is good advertising for Musk’s brand. One other possible benefit of the lawsuit to Musk could very well be the forcing of OpenAI’s hand to make its products available to everyone, and therefore unprofitable, thereby eliminating Musk’s competition.
Musk has his own AI company, x.AI, which introduced Grok as competition to ChatGPT late last year, a fact that OpenAI raised in its response to Elon Musk. Early this month, OpenAI discussed the situation with the public in an official blog post, releasing several of Musk’s emails with OpenAI executives, back when he was a member of the organisation. These emails suggest that Musk is among the primary proponents of the commercialisation of AI, arguing that the platform stood no chance of success if it relied only on raising cash. As a stand-alone statement, this is of course factually correct, considering the amount of dollars that have gone towards building generative AI to where it is today. According to the blog, in the company’s early days, OpenAI had been contemplating raising $100 million to fund its activities. Musk has allegedly responded to these plans in an email to Altman and Brockman stating that “we need to go with a much bigger number than $100 (million) to avoid sounding hopeless… I think we should say that we are starting with a $1B funding commitment… I will cover whatever else anyone else doesn’t provide.”
However, according to OpenAI, Musk never followed up on this promise and only ever committed $45 million in funding for OpenAI. Other donors on the other hand had raised $90 million. Musk’s lawyers are yet to comment on these accusations. In another email, Musk had tried to convince OpenAI executives that the only viable way for the company to move forward was to be merged into Tesla or failing that, handing over total control to him. As quoted by OpenAI, Musk said that the company should “attach to Tesla as its cash cow” as “Tesla is the only path that could even hope to hold a candle to Google. Even then, the probability of being a counterweight to Google is small. It just isn’t zero.”
If anything is obvious from the outside, it’s just that both these entities, OpenAI and Musk have their interests at heart in this dispute. It is impossible that Musk and his legal team are unaware that their lawsuit has little potential in court without an actual breached contract. OpenAI also has little reason beyond the optics of the situation to take to a blog post to defend its operations. Notably, it is also doing very little to prove its disinterest in profit to the public, instead stating that profits are a necessary part of developing the technology behind AI going forward. Arguably, the ethics of commercialised generative AI products bears the least relevance in the legal proceedings that are to follow, if they make it into trial at all.
(Theruni M. Liyanage)