A group of former open employees on Friday testified one Suggested short amicus In support of Elon Musk in his lawsuit against Openai, opposed to the scheduled conversion of Openai from a non -profit to a profit company.
The short, filed by Harvard’s law professor and the founder of Creative Commons Lawrence Lessig, twelve former Openai employees: Steven Adler, Rosemary Campbell, William Chowdhury, Jacob Hilton, Daniel Kokotajlo, Gretchen Krrueger, Todor Markov, Richard Ngo, Girish Sasry, William Sakunders, Carrol Wainwright, Jeffright, Jeffright, Jeffright Waffright. It makes the assumption that if OpenAi’s non -profit control has given control of the organization’s business activities, it will “fundamentally violate its mission”.
Many of the former executives have spoken against Openai’s practices publicly before. Cruise has called the company to improve accountability and transparency while Kokotajlo and Saunders previously warned that Openai is in a “Reckless” race for AI domination. Wainwright said This Openai ”should not [be trusted] When he promises to do the right one later. ”
In a statement, a spokesman for Openai said that the Openai non -profit organization “does not go anywhere” and that the organization’s mission “will remain the same”.
“Our council was very clear,” the spokesman told TechCrunch by email. “We are turning the existing speculative arm into a public benefit company (PBC)-the same structure as other AI laboratories such as Anthropic-where some of these former employees are now working-and [Musk’s AI startup] xai. ”
Openai was founded as a non -profit in 2015, but turned into a “capped speculative” in 2019 and is now trying to restructure once again into a PBC. When it was transferred to a profit with coverage, Openai maintained its non -profit wing, which today has a share of control over the corporate arm of the organization.
Musk’s costume against Openai accuses the launch of its non -profit mission, which aimed to secure AI research, benefits from all over humanity. Musk had requested a preliminary order to convert Openai. A federal judge denied the request, but allowed the case to go to a jury trial in the spring of 2026.
According to the brief reference to the former open workers, today’s OpenAI structure-a non-profit controlled group of other subsidiaries-is a “critical part” of its overall strategy and “critical” of the organization’s mission. The restructuring that abolishes the control of the non -profit organization would not only contradict the commitments of the mission and charter of Openai, but would “violate the confidence of workers, donors and other stakeholders who have joined and supported the organization on the basis of these bindings”.
‘Openai pledged several basic principles for execution [its] Mission to the map of the map, “the brief report said. The court should recognize that maintaining the governance of the non -profit organization is essential to maintain the unique structure of Openai, which was designed to ensure that the artificial general benefit of humanity is beneficial.”
Artificial general intelligence, or agi, is widely understood that it means AI that can complete a human container.
According to short, Openai often used its structure as a recruitment tool – and repeatedly guaranteed staff that non -profit control was “critical” in performing its mission. Shortly reports an Openai All Hands meeting by the end of 2020 during which Openai Sam Altman CEO allegedly emphasized that the governance and supervision of non -profit organizations were “primary” in “securing that security and widespread social benefits were prioritized.
“When hiring talks with candidates, it was common to mention the only OpenAI governance structure as a critical difference between OpenAi and competitors such as Google or the man and an important reason they should consider joining the company,” the brief said. “This same reason was also often used to persuade employees who were thinking of leaving for competitors to stay in Openai – including some of us.”
Shortly warns that, it should be allowed in Openai to turn into a speculative, could be encouraged to “[cut] Angles “for security work and develop a strong” concentrated between its shareholders “. A speculative Openai will have no reason to comply with the” Merge and Help “clause on the current map of Openai, which binds that Openai will stop competing with any” the brief report.
Former open employees, some of whom were research and policy leaders in the company, participated in a growing group that expresses a strong opposition to the transition to Openai.
Earlier this week, a group of organizations, including non -profit and working groups, such as California Teamsters, asked California Attorney Rob Bonta to stop Openai to become a profit. They claimed that the company “failed to protect its charity” and actively overturning its charity mission to promote safe artificial intelligence. “
ENCODE, a non -profit organization that, at the same time, congratulated California’s unusual SB 1047 security legislation, said similar concerns in a brief report submitted in December in December.
Openai said its conversion would maintain the non -profit arm and put it on with resources that would be spent on “charity initiatives” in areas such as health care, education and science. In return for controlling its involvement in Openai’s business, the non -profit company would have been reported to earn billions of dollars.
“We’re really ready to build the best-equipped non-profit organization that the world has ever seen-we don’t turn it away.” The company wrote in a series of posts in x Wednesday.
The stakes are high for OpenAI, which must complete the conversion of the speculative nature by the end of this year or next or risk to resign from some of the funds that have increased in recent months, according to reports.
