oh my, i have nothing to say but going to sit here for the comments.

edit: actually I can't sit it out. To some extent what you are suggesting is probably already what companies will do.

Saying that your version is just a tad more unethical and has probably already been done for years with AI chatbots before LLMS.