Novelists sue OpenAI for scraping their books to train ChatGPT (opens in new tab)

(theregister.com)

2 pointsPseudocrat2y ago2 comments

2 comments

It's early days in AI copyright lawsuits, but IMHO this won't be successful - one doesn't need a license to produce a summary of a book.

Does Wikipedia need a license to publish an article about a book, or does a schoolkid need a license to write his homework about a book?

It was interesting to me that OpenAI admitted (according to the article) that they used Bibliotik to train the models.

j / k navigate · click thread line to collapse

It's early days in AI copyright lawsuits, but IMHO this won't be successful - one doesn't need a license to produce a summary of a book.

Does Wikipedia need a license to publish an article about a book, or does a schoolkid need a license to write his homework about a book?

It was interesting to me that OpenAI admitted (according to the article) that they used Bibliotik to train the models.

j / k navigate · click thread line to collapse