Elon Musk, one of the co-founders of OpenAI, played a significant role in the company’s early days, when it was still a simple startup. Just recently, he had remembered the early days of OpenAIwith the donation – in 2016 – of that historic NVidia DGX-1 server which marked the beginning of the development of modern generative models and applications based onartificial intelligence. “Look what OpenAI has become today“, Musk wrote just a few days ago, blatantly contesting the departure from the company’s founding principles.
Musk contributed financially to the founding of OpenAI and actively supported the organization’s mission. When OpenAI was founded in 2015, the idea was to promote and develop artificial intelligence in a ethicalso that i benefits would reverberate across all of humanity.
The legal dispute initiated against OpenAI and Sam Altman
According to Musk, OpenAI and, in particular, the current CEOs Sam Altman and president Gregory Brokman, would have violated the initial agreements moving away from the initial mission of the company. The accusatory thesis is that OpenAI has profits given priority compared to the platform’s original goal, which was to serve the interests of humanity.
The complaint stems from Musk’s belief that OpenAI has moved away from its initial approach, putting revenues and economics above the ethical considerations and social benefits that AI developments should bring.
“OpenAI has transformed into a de facto subsidiary, a closed codeof the largest technology company in the world: Microsoft. With his new board of directors, he is not just developing, but actually perfecting a general artificial intelligence to maximize Microsoft’s profits, rather than for the good of humanity“, Musk claims in his complaint.
Musk’s allegations include alleged breaches of contract, breaches of fiduciary duty and unfair business practices. The entrepreneur asks that OpenAI return to being an open reality, based on philosophy open source and also requests an injunction to prevent the company’s management and Microsoft from profiting from the artificial intelligence technology developed by the company.
Finger pointing at the GPT-4 model
The idea of models GPT (Generative Pre-trained Transformer) of OpenAI has its roots in the concept of transformer, illustrated in 2017 by a team of Google software engineers.
Musk claims that the GPT-4 model released in March 2023 is a closed tool, unlike previous iterations, a choice driven by commercial considerations rather than general interests.
“The internal details of GPT-4 are known only to OpenAI and (…) Microsoft. GPT-4 is therefore the opposite of ‘open AI’“, Musk says in the complaint. “The model is closed for proprietary commercial reasons: Microsoft will make a fortune selling GPT-4 to the public, which would not be possible if OpenAI, as it is required to do, freely made the technology available to global users“.
In his complaint, Musk also addresses the issue of Altman’s dismissal and subsequent reinstatement as CEO. According to Musk, Altman’s departure prompted Microsoft to intervene and force the resignations of board members who tried to remove him. The current members of the board, Musk tries to point out, are no longer scientists and researchers who support and understand the technology.
Musk, who initially sat on OpenAI’s board of directors himself and left the company in 2018, says the conflict between management and Altman exacerbated following the development of GPT-4 and ahead of the development of aAGI (Artificial General Intelligence), with all the related ethical and safety issues. We spoke about the new technology, probably called OpenAI Q-Star (Q*), some time ago with quite a bit of skepticism.