Over the last 24 hours, unsettling developments within OpenAI have unraveled, with a series of high-profile exits spinning the AI research organization into crisis.
The disruptions began with Sam Altman, co-founder and CEO, who participated in the APEC Business Summit and a Burning Man event last week. On Friday, he received a text message from Ilya Sutskever, co-founder and Chief Scientist, followed by a virtual meeting where three other independent directors, apart from Brockman, were present. Altman was informed by Sutskever that he was fired, with an official statement indicating his removal for being dishonest to the board.
Greg Brockman, co-founder, chairman, and president, also received a Google Meet invitation shortly after, where he was informed of his removal from the board. Despite his importance to the company, he was retained as president but will now report to interim CEO Mira Murati, who knew beforehand that she would succeed Altman.
The sudden upheavals instigated a mass resignation within OpenAI. Three senior figures, including company director of research Jakub Pachocki, potential AI risks assessment team leader Aleksander Madry, and Szymon Sidor, a researcher with seven years of tenure, resigned immediately. More resignations are expected.
The board's decision to fire Altman without a full board meeting and before Brockman, the chairman, was informed marks a significant governance concern, showcasing the board's lack of experience in managing such a critical company.
The root of the trouble seems to lie in the conflicts between prioritizing safety and profitability within OpenAI. Altman, leading the "for-profit faction," advocated for commercial applications, while Sutskever underscored safety and the underlying principles of OpenAI, representing the "non-profit faction."
Moreover, Altman's attempts to raise funds and consider potentially unsuitable governance models for new businesses have reportedly dissatisfied Sutskever and his colleagues.
Albeit the detrimental impact of these resignations, the potential of generative AI goes beyond developing more secure, advanced models. The revolution generative AI kickstarted is still in its early phase, and OpenAI's mission to research safe general artificial intelligence might see a change in direction. The commercial potential Altman envisioned for OpenAI might depart, but the organization would continue to require significant investment to sustain its research. As observed, OpenAI's value might not lie in becoming another tech giant but in its ability to unlock the potential of generative AI for countless researchers, developers, and users.