OpenAI unveils new safety board after employee revolt

OpenAI announced on Tuesday the formation of a new committee dedicated to making safety and security recommendations to the company’s board. This move comes weeks after OpenAI disbanded a team focused on AI safety.

In a blog post, OpenAI revealed that the new committee will be led by CEO Sam Altman, board chair Bret Taylor, and board member Nicole Seligman.

This development follows the recent departure of Jan Leike, an OpenAI executive who focused on safety. Leike resigned, criticizing the company for underinvesting in AI safety and highlighting growing tensions with OpenAI’s leadership. Additionally, Ilya Sutskever, a key figure in OpenAI’s “superalignment” team, also left. Sutskever had previously played a pivotal role in Altman’s unexpected removal as CEO, only to later support his reinstatement.

Earlier this month, an OpenAI spokesperson informed CNN that disbanding the superalignment team and redistributing its members throughout the company would better align with its superalignment objectives.

In the blog post, OpenAI also mentioned it has started training a new AI model to succeed the one currently powering ChatGPT. This new model, which will follow GPT-4, is seen as another step toward achieving artificial general intelligence.

“We are proud to develop and release models that lead the industry in both capabilities and safety, and we welcome a robust debate during this critical time,” the company stated.

The initial task of the Safety and Security Committee is to assess and enhance OpenAI’s processes and safeguards over the next 90 days. At the end of this period, the committee will present their recommendations to the full Board. After the Board’s review, OpenAI will publicly share an update on the adopted recommendations in a manner consistent with safety and security.