OpenAI declares AI race “over” if training on copyrighted works isn’t fair use - Ars Technica

As a scifi nerd, I doubt these guys actually read scifi.

We've also already gotten some benefit for truely useful, non-copyright infringing AI (generally the kinds used by scientists evaluating big data sets like protein folding and finding exoplanets). None of those AIs rely on stolen art or stolen copyrighted material. But they have benefited somewhat from the pursuit of AI processing hardware. So even if you are pro-AI where it has shown any real value as a tool, you still don't need the copyright infringement. Let's kill OpenAI, and other LLMs that stole to get where they are. The actually useful AI probably won't go anywhere, even if the companies pushing their AI the most right now all go under.

You said that well. He always looks so disturbing.