The OpenAI chief executive, Sam Altman, is facing accusations of creating a toxic culture of lying and engaging in psychological abuse within the company, according to two former board members. Helen Toner and Tasha McCauley, who were part of a failed attempt to oust Altman in November, stand by their decision to remove him due to his alleged behavior undermining key decisions and internal safety protocols.
These accusations shed light on the challenges faced by OpenAI, the company behind the powerful AI chatbot ChatGPT, which aims to develop safe AI technology. Despite efforts to position the company as a leader in safe AI development, concerns have been raised about Altman’s leadership and the company’s governance.
The departure of Ilya Sutskever, a co-founder of OpenAI who initially supported removing Altman from the board, further emphasizes the internal strife within the company. The recent resignation of senior safety researcher Jan Leike, citing a lack of focus on safety culture, adds to the growing concerns about OpenAI’s direction.
Ms. Toner and Ms. McCauley have called for government regulation of AI businesses, arguing that self-regulation may not be sufficient, especially in the face of profit incentives. They warn that recent developments within OpenAI, including Altman’s reinstatement and the departure of safety-focused talent, raise doubts about the company’s commitment to safe AI development.
Despite a brief period of Altman’s removal from the board, he quickly returned as chief executive with the support of staff and investors. A new board, free of his critics, was established, raising questions about governance and accountability within the company.
The reasons behind the failed ousting of Altman remain unclear, with the board only acknowledging his lack of consistency in communication. An external review found that Altman’s behavior did not warrant his removal, leading to his reinstatement.
As OpenAI navigates these internal challenges, the future of the company’s self-regulation and commitment to safe AI development remain uncertain. The ongoing debate around Altman’s leadership and governance of the company highlights the complexities of overseeing AI technology in a rapidly evolving industry.
