News for hackers. Stuff that matters.

A comment on OpenAI now tries to hide that ChatGPT was trained on copyrighted books. Parent comment

Re:OpenAI now tries to hide that ChatGPT was trained on copy... by jameshart on Thursday August 24, 2023 @10:43PM (#37257355)
I was trained on copyrighted books. I only get in trouble when I spout out paragraphs from them from memory and pass them off as my own.

I know not to do that, though.

Seems only fair that the same should apply to GPT.

[ Parent | on HN ]
Re:OpenAI now tries to hide that ChatGPT was trained on copy... by ZeroZeroOneZero on Thursday August 24, 2023 @11:36PM (#37257725)
Agreed.

Honestly, this is starting to feel like copyright holders using the Big, New, Scary AI as a strawman to attack Fair Use.

[ Parent | on HN ]