OpenAI forms team to oversee ‘superintelligent’ AI but fails to maintain it, according to insider

OpenAI’s Superalignment team, tasked with overseeing the development of “superintelligent” AI systems, has been rocked by resignations this week. Co-lead Jan Leike, a key figure in the team and a former DeepMind researcher, publicly announced his departure citing disagreements with OpenAI leadership over the company’s priorities.

Leike expressed concerns about the lack of focus on critical areas such as security, monitoring, safety, and societal impact, which he believes are essential for the responsible development of advanced AI models. His departure, along with that of OpenAI co-founder Ilya Sutskever, has raised questions about the company’s commitment to its mission of developing AI for the benefit of humanity.

The Superalignment team was formed last year with the goal of addressing the technical challenges of controlling superintelligent AI. Despite making significant contributions to safety research and funding external projects, the team faced obstacles in securing the resources needed to fulfill its mission.

Following the resignations, OpenAI has restructured its approach to AI development, with John Schulman taking over the responsibilities of the Superalignment team. However, concerns remain about the company’s shift away from a dedicated safety-focused team.

As OpenAI navigates these changes, the future of its AI development efforts and the potential impact on society remain uncertain. The departures of key team members highlight the challenges of balancing innovation with ethical considerations in the rapidly evolving field of artificial intelligence.