|
|
|
|
|
by AnimalMuppet
15 days ago
|
|
And the ones spanning generations were completely fair game. The ones that were still under copyright are a different matter. (It has not, so far as I know, been legally decided, but it's probably going to depend on whether AI is a transformative use.) |
|
Given the sheer volume of information posted to the Internet in the last 40-50 years, I'd wager that covers 80% or more of the relevant input data.
Old text is relatively scarce in the grand scheme of things.
But I have no real clue, just spitballing.