Musk claimed OpenAI and Altman elected to use the company’s technology “to maximize profits” for Microsoft rather than for the benefit of humanity, adding that its entire line of development “is now veiled in secrecy.”
Topline
Elon Musk sued ChatGPT-maker OpenAI and its CEO Sam Altman Thursday for undermining the company’s original goal of developing an open-source artificial general intelligence that would “benefit humanity” in favor of maximizing profits, widening a longstanding rift between the two cofounders over the future of a company that now sits at the heart of Silicon Valley and the AI revolution.Key Facts
In court documents filed with the San Francisco Superior Court, Musk claimed Altman and Greg Brockman, another OpenAI cofounder, approached him in 2015 to launch a non-profit organization with a goal of developing AI “for the benefit of humanity” rather than to enrich shareholders.
Its work would rival leading AI labs like Google/DeepMind and, barring safety concerns, be open-source—freely and publicly available for anyone to use, modify and distribute—principles Musk said were then enshrined in OpenAI’s founding agreement.
Altman, Brockman and OpenAI have since reneged on that agreement by restricting access to its technology in order to “maximize profits,” Musk’s lawyers alleged, claiming breach of contract, breach of fiduciary duty and unfair competition and asking the court to force it to return to open source and prevent the company and named founders , as well as backer Microsoft, from profiting off the technology.
OpenAI’s close relationship with Microsoft, which has invested billions into the venture and makes use of its AI models across its products, illustrates this pivot to closed-source profit seeking, Musk claims, with secrecy over technology now “primarily driven by commercial considerations, not safety.”
The billionaire claimed Microsoft “stands to make a fortune” selling the company’s most recent AI model, GPT-4, to the public, slamming it as a “de facto Microsoft proprietary algorithm” and the “opposite of ‘open AI’” and something that would “not be possible” if OpenAI made its technology freely available to the public “as it is required to do.
