OpenAI releases former employees from NDAs, making non-disparagement agreements unenforceable

From CNBC: 2024-05-23 23:05:00

OpenAI has decided to revoke a requirement for former employees to choose between signing a non-disparagement agreement or forfeiting vested equity. An internal memo to ex-workers confirmed that vested units will not be retracted, and any non-disparagement contracts are unenforceable. OpenAI is updating their departure process and values amid controversy.

In response to mounting criticism, OpenAI withdrew an audio voice for ChatGPT named “Sky,” which resembled Scarlett Johansson’s voice without her consent. The organization is pausing the use of Sky and addressing voice selection in ChatGPT. Additionally, OpenAI disbanded its long-term AI risks team, with members reassigned to other departments, following leaders’ departures.

OpenAI’s Superalignment team, focused on steering and controlling smarter AI systems, saw key members depart while the company faces scrutiny over its safety culture. Despite challenges, co-founder and CEO Sam Altman expressed a commitment to raising awareness of AGI risks and opportunities. This shift reflects ongoing challenges for OpenAI as it navigates internal and external changes.



Read more at CNBC: OpenAI sends internal memo releasing former employees from NDAs