The safety committee will exercise oversight over model launches, including having the authority to delay a release until safety concerns are addressed, as stated in a recent blog post.
The committee was first formed in May, following a series of high-level departures who publicly worried that the ChatGPT maker could not govern itself and was no longer acting responsibly about artificial intelligence.
Bret Taylor, OpenAI's chair, has stepped down from the committee, providing relief to critics who questioned if the committee could serve as a watchdog with a CEO present.
OpenAI's troubles heightened last month when the company battled to stop an AI safety bill in California, claiming it would stifle progress and push companies out of the state.
[
Collection
]
[
|
...
]