The biggest problem IP and Copyright laws face is AI, every data accessible on the Internet is already been used to train these models, you might be the owner of a robust dataset but an AI lab somewhere in the world right now might br training their model with your data for free.
This is why I built Modelfuel for the @StoryProtocol CDR hackathon, Dataset owners can list their datasets for whatever they valued it at and choose how long the listing should stay active, they can also delist at any time.
Buyers ( AI labs, organizations or Individuals) can buy the license to your dataset and train their models, but this time you can actually make money off your Data.
Using CDR sellers can upload their encrypted data to CDR vaults, buyers can buy the PIL and once confirmed onchain, they get access to the decrypted dataset and can download directly to their local machine
Repo and website below