Sam Altman ‘genuinely embarrassed’ by OpenAI’s harsh NDAs

On Friday, following the departure of several high-level OpenAI employees, a Vox article revealed that OpenAI employees are required to sign a restrictive severance agreement. Now, OpenAI CEO Sam Altman says part of the deal should never have happened and that it’s “one of the few times I’ve been truly embarrassed leading OpenAI “.

The agreement includes both nondisclosure and nondisparagement provisions that prohibit departing employees from criticizing OpenAI for life; they can’t even say they signed an NDA. Failure to sign the agreement or breaking it means the former employee loses all of their vested shares in the company.

“We have never recouped anyone’s acquired equity, nor will we if people do not sign a separation agreement (or agree to a non-disparagement agreement). Equity earned is equity earned, period,” Altman said. in a Saturday post on X.

“There was a provision for potential reversal of actions in our previous release documents; even if we never got anything back, this should never have been something we had in any document or communication. It is my fault and one of the few times I’ve been really embarrassed to use OpenAI I didn’t know this was happening and should have done it.”

Altman says OpenAI is revamping its release documents. “If any former employee who signed one of these old agreements has any concerns, they can contact me and we will resolve that as well. I am truly sorry.”

This Tweet is currently unavailable. It may be loading or has been deleted.

OpenAI researcher Jan Leike resigned from OpenAI this week alongside fellow researcher Ilya Sutskever. Leike said he “finally reached a breaking point” on the company’s core priorities, although Sutskever said he believed the company was on the right path to developing artificial general intelligence. safe and beneficial.

OpenAI unveils its ChatGPT AI voice assistant