4 ms·
I can kind of see that, but he's just one person and he bought the books he learned from. I presume OpenAI doesn't pay a fraction of the book's cost up to the
by gigel82 4y ago
I can kind of see that, but he's just one person and he bought the books he learned from.
I presume OpenAI doesn't pay a fraction of the book's cost up to the writer every time it spits out content based on the book (it -at most- paid for the book once and ingested all its content to be used forever).
- brookst 4y agoI don't think OpenAI is trained on any copyrighted, for-sale books. GPT-3 at least uses books1 and books2. Books1 seems to be Bookcorpus[0], Books2 seems to be Libgen[1]. There is some unclarity about the purity of licensing in Bookcorpus affecting 200-400 of the 11,000 books[2] 0. https://paperswithcode.com/dataset/bookcorpus https://paperswithcode.com/dataset/bookcorpus 1. https://librarygenesis.pro/ https://librarygenesis.pro/ 2. https://towardsdatascience.com/dirty-secrets-of-bookcorpus-a-key-dataset-in-machine-learning-6ee2927e8650 https://towardsdatascience.com/dirty-secrets-of-bookcorpus-a... Are you suggesting that free / public domain material shouldn't be used for training, or were you not familiar with the training set?