I guess to state what I think should be a real solution to this problem: people should have their own AI lawyer that works on their behalf to flag things they m

I guess to state what I think should be a real solution to this problem: people should have their own AI lawyer that works on their behalf to flag things they might want to object to. I helped build a surprisingly effective prototype of such a thing using a very simple "bag of words" approach, and it had >99% accuracy in identifying clauses, because they almost all use the same words, basically by design. So this is pretty trivial to build in the grand scheme of things. Courts should prevent companies from banning AI assistants, that is where they could help.