OpenAI now tries to hide that ChatGPT was trained on copyrighted books, including J.K. Rowling’s Harry Potter series::A new research paper laid out ways in which AI developers should try and avoid showing LLMs have been trained on copyrighted material.
The problem with trying to restrict AI from scanning the art and making conclusions about it is that it’s the same as trying to ban humans from creating art that’s inspired by other art. It’s the same process even. If the AI is actually producing one-for-one copies of their work, you might have a leg to stand on in terms of arguing the AI shouldn’t be compensated for creating those specifically, but it’s creating works that are just loosely influenced by seeing the original art.