Looks like they need to poison the data.
Training is fair use, according to the courts. They can demand all they want but it’s over.
They seem to be referencing generated content, not training.
The Content Overseas Distribution Association (CODA), an anti-piracy organization representing Japanese IP holders like Studio Ghibli and Bandai Namco, released a letter last week asking OpenAI to stop using its members’ content to train Sora 2
This is literally the first sentence of the article.
Wow, now read the second paragraph.
Sora 2 generated an avalanche of content containing Japanese IP after it launched on September 30th, prompting Japan’s government to formally ask OpenAI to stop.
My point is that it’s a totally different legal avenue. Even if they can legally train on it, they cannot generate and distribute content which violates IP.
Fair enough! It’s certainly the right avenue to attack the copyright violations from.
