Comment by Zvi Mowshowitz

OpenAI has three less good safety researchers. Trust will be lower. Speech will be chilled. [...] The only way to affirm an open and transparent culture of dialogue, and okay with full collaboration with outsiders and auditors and evaluators on safet...

I do think we should do that work, and give it, and its details, serious consideration.

I am not endorsing the vision, courses of action suggested here or the predictions made, but I am endorsing 1. the idea that this is well-written, carefully considered, and says what those writing it sincerely believe, and that 2. you should read it ...