OpenAI has disbanded its team dedicated to long-term risks associated with artificial intelligence just one year after its formation, as confirmed by a source familiar with the situation to CNBC on Friday.
The source, speaking anonymously, mentioned that some team members are being reassigned to various other teams within the organization.
This development follows the recent departures of both team leaders, OpenAI co-founder Ilya Sutskever and Jan Leike, from the Microsoft-backed startup. Leike expressed concerns on Friday about OpenAI’s shift in focus from safety to product development.
OpenAI did not respond immediately to CNBC’s request for comment.
Established last year, OpenAI’s Superalignment team aimed to achieve scientific and technical breakthroughs in guiding and controlling AI systems that are more intelligent than humans. The initiative was initially set to receive 20% of the company’s computing power over a four-year period.
Wired was the first to report on the team’s dissolution.
Sutskever and Leike announced their exits on social media platform X earlier this week, with Leike elaborating on his reasons for leaving the company on Friday.
Leike expressed his belief that OpenAI should prioritize security, monitoring, preparedness, safety, and societal impact more intensively. He emphasized the importance of focusing on these areas to ensure responsible development of AI technology.
Leike stressed the need for OpenAI to become a “safety-first AGI company” due to the inherent risks associated with building machines smarter than humans. He highlighted his concerns about the company’s current trajectory and its lack of emphasis on safety measures.
Leike did not provide an immediate response to a request for comment.
These high-profile departures come in the wake of a leadership crisis at OpenAI involving co-founder and former CEO Sam Altman.
Last year, Altman was removed from his position by the company’s board, citing communication issues. The situation escalated, with reports indicating conflicting priorities within the organization regarding AI development.
The upheaval led to resignations and tensions among employees and investors, including Microsoft. Altman was eventually reinstated following the turmoil.
Altman expressed sadness over Sutskever’s departure, praising his contributions to the field of AI and announcing Jakub Pachocki as Sutskever’s successor as chief scientist.
News of the team disbandment and the departures of Sutskever and Leike coincided with OpenAI’s recent launch of a new AI model and desktop version of ChatGPT, signaling the company’s continued efforts to enhance its chatbot technology.
The updated GPT-4 model, known as GPT-4o, boasts improved speed and capabilities in text, video, and audio interactions, with plans to introduce video chat functionality in the future.
