The cure for most of the problems of Large Language Model generative AI systems is making the firms absolutely liable for what their AI models say and do. Financially, Legally, and Criminally. The firms can then decide which of their AI services they wish to continue deploying to the public under that set of constraints. It will be completely up to them!