[bee_rider]

“Not possible” = “a business-destroying level of honesty”?

[rcxdude]

Even if training on the copyrighted material is OK, just providing a data dump of it almost certainly is not.

[alpaca128]

No need for a data dump, just list all URLs or whatever else of their training data sources. Afaik that's how the LAION training dataset was published.

[anonymoushn]

providing a large list of bitrotted URLs and titles of books which the user should OCR themselves before attempting to reproduce the model doesn't seem very useful.

[echoangle]

Aren't the datasets mostly shared in torrents? They probably won't bitrot for some time.

[Wowfunhappy]

...no? They also use web crawlers.

[bee_rider]

The datasets are collected using web crawlers, but that doesn’t tell us anything about how they are stored and re-distributed, right?

[tokioyoyo]

There is a "keep doing what you're doing, as we would want one of our companies to be on top of the AI race" signal from the governments. It could've been stopped, maybe, 5 years ago. But now we're way past it, so nobody cares about these sort of arguments.