OpenAI wanted to preach safety. I think we hold them liable for literally everything, everything Chat GPT does or says, until the model is open and they can argue that they have no control over it.

They wanted this liability, they accepted this liability, they said they'd keep it safe and they haven't. It's on them.