Employees of OpenAI Who Are Key Pillars of Safety Are Leaving the Company

Citation: Image used for information purpose only. Picture Credit: https://scx2.b-cdn.net

Recently, a number of important members of Sam Altman’s AI startup’s Super-alignment team—which was meant to prevent AI from going awry—went missing.

The ultimate objective of OpenAI’s Super-alignment team, which was established about a year ago, is to maintain control over hypothetically superintelligent AI systems and keep them from turning against people. Naturally, a lot of people were worried. After all, why was there a need for a squad like this in the first place? More worrying news has already surfaced: Ilya Sutskever and Jan Leike, the team leaders, recently left OpenAI.

Following a string of high-profile exits from the company, including those from Sutskever and Leike’s safety-focused team, Superalignment’s leadership has resigned. Sutskever and the OpenAI board spearheaded a botched attempt to remove CEO Sam Altman back in November 2023. After six months, a number of OpenAI employees who were involved in important safety teams or who made public statements regarding AI safety have departed the business.

In the end, Sutskever issued an apology for the coup (my mistake, buddy!) and joined 738 other OpenAI workers (out of 770 total) in a letter requesting the reinstatement of President Greg Brockman and Altman.

Sources claim that a number of the now-resigned employees either neglected to sign the statement endorsing OpenAI’s leadership or did so slowly.

The names of Jan Leike, Leopold Aschenbrenner, and William Saunders, members of the Superalignment team who have subsequently resigned, are absent from the Times edition along with the names of over 700 additional OpenAI employees who have expressed support for Altman and Brockman. Not included in this early draft of the letter are prominent AI researcher Andrej Karpathy, as well as former employees Daniel Kokotajlo and Cullen O’Keefe, both of whom have since departed OpenAI. These people might have been the last to sign the updated version of the letter, but they certainly don’t appear to have done so.

More generally, there has always been disagreement over safety at OpenAI. This is the reason why Dario and Daniela Amodei, together with nine other former OpenAI employees, founded their own AI startup, Anthropic, in 2021. It was also alleged that the nonprofit board members of OpenAI removed Altman and Brockman due to safety concerns. These board members were swapped out by a few notorious Silicon Valley businesspeople.

Read More : https://thecioleaders.com/