Is OpenAI ready for AGI? Not so much anymore now

OpenAI has decided to dissolve its AGI Readiness team, which was tasked with assessing whether the company is prepared to handle an advanced artificial intelligence and if our society could manage such a powerful technology. Miles Brundage, the principal advisor of this team, announced his departure through a statement on Substack, where he explained that the decision was due to the desire to adopt a more impartial approach. “I thought my research would have more impact outside of OpenAI,” he stated.

General artificial intelligence (AGI) is a field of AI that seeks to create systems that can match or surpass human intelligence in various tasks. This topic has generated a great debate both socially and in the technological world, with positions ranging from those who believe that AGI is close to becoming a reality, to those who claim it will never be achieved.

Brundage expressed in his message that “neither OpenAI nor any other lab is prepared, and the world isn’t either”. He also revealed his intention to found a non-profit organization or join an existing one to focus on research policies in the field of AI. “It is unlikely that AI will be as safe and beneficial as possible without a concerted effort,” he commented, emphasizing the need for a collective approach to optimize the benefits and minimize the risks of AI.

The former members of the AGI Readiness team will be reassigned to other teams within the company, while OpenAI expressed its support for Brundage’s decision, expressing gratitude for the advisor’s contributions and valuing his dedication to independent research.

This change follows the dissolution in May of OpenAI’s Superalignment team, which focused on the development of advanced control systems for AI, and comes after the recent departures of other senior executives from OpenAI, including Chief Technology Officer Mira Murati and research leads Bob McGrew and Barret Zoph.