According to the recent resignations, OpenAI has decided to disband the “Superalignment” team and incorporate its operations into other internal research initiatives. Every member of the OpenAI team that examined the existential risks posed by AI has apparently left or been taken over by other independent research teams.
The other co-lead of OpenAI’s super alignment team, Jan Leike, was a former DeepMind researcher who announced his resignation a few days after Ilya Sutskever, the company’s chief scientist and one of its co-founders, made his announcement on X. As to Leike’s statement, he left the company because he felt that its priorities prioritised product development over AI safety.
Leike wrote a series of essays arguing that as Artificial General Intelligence (AGI) development progresses, safety and readiness should take precedence over other considerations, and that the OpenAI leadership made a mistake in selecting their key priorities.
The phrase artificial general intelligence (AGI) refers to a theoretical intelligence that can accomplish a variety of jobs either as well as or better than humans. Leike left OpenAI after three years, accusing the company of putting more emphasis on creating eye-catching products than fostering a strong AI safety culture and procedures.
He underlined how critical it was to allocate resources—especially computer power—immediately in order to support his team’s critically important but underfunded safety study. ‘We finally came to a breaking point since I had been at odds with OpenAI leadership for quite some time on the company’s primary aims. Over the past few months, my team has been sailing against the wind”
To get ready for the arrival of artificial intelligence that is so sophisticated that it could eventually surpass and surpass its creators, OpenAI established a new research team in July of last year. This new team was given 20% of OpenAI’s computational resources, and its principal scientist and co-founder, Ilya Sutskever, was named co-leader.
Owing to the recent resignations, OpenAI has decided to disband the “Superalignment” team and incorporate its operations into other internal research initiatives. The current internal reorganisation, which was started in reaction to the governance issue in November 2023, is purportedly the reason behind this move.
Sutskever was a part of the team that was successful in getting Altman rehired as CEO of OpenAI after he was briefly removed by the board in November of last year due to criticism from staff. The Information claims that Sutskever told staff members that the board’s choice to dismiss Sam Altman satisfied their obligation to ensure that OpenAI creates AGI that is beneficial to all people. Sutskever, one of the six board members, highlighted the group’s dedication to coordinating OpenAI’s objectives with the larger good.