I would be very happy if either a court or lawmakers decided that copyright itself was unconscionable. That isn't what's going to happen, though. And I think it's incredibly unacceptable if a court or lawmakers instead decide that AI training in particular gets a special exception to violate other people's copyrights on a massive scale when nobody else gets to do that.
As far as a fair use argument in particular, fair use in the US is a fourfold test:
> the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
The purpose and character is absolutely heavily commercial and makes a great deal of money for the companies building the AIs. A primary use is to create other works of commercial value competing with the original works.
> the nature of the copyrighted work;
There's nothing about the works used for AI training that makes them any less entitled to copyright protections or more permissive of fair use than anything else. They're not unpublished, they're not merely collections of facts or ideas, etc.
> the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
AI training uses entire works, not excerpts.
> the effect of the use upon the potential market for or value of the copyrighted work.
AI models are having a massive effect on the market for and value of the works they train on, as is being widely discussed in multiple industries. Art, writing, voice acting, code; in any market AI can generate content for, the value of such content goes down. (This argument does not require AI to be as good as humans. Even if the high end of every market produces work substantially better than AI, flooding a market with unlimited amounts of cheap/free low-end content still has a massive effect.)