One hack like this attacking OpenAI using an open model would be devastating to the mission of openness -- potentially ending. The double standard on safety in the AI industry is pretty bizarre.
"But china" will come in too, as it'll likely be a Chinese model
