Comment by TeMPOraL
Comment by TeMPOraL 13 hours ago
Well, so can a nontrivial number of people. It's Harry Potter we're talking about - it's up there with The Bible in popularity ranking.
I'm gonna bet that Llama 3.1 can recall a significant portion of Pride and Prejudice too.
With examples of this magnitude, it's normal and entirely expected this can happen - as it does with people[0] - the only thing this is really telling us is that the model doesn't understand its position in the society well enough to know to shut up; that obliging the request is going to land it, or its owners, into trouble.
In some way, it's actually perverted.
EDIT: it's even worse than that. What the research seems to be measuring is that the models recognize sentence-sized pieces of the book as likely continuations of an earlier sentence-sized piece. Not whether it'll reproduce that text when used straightforwardly - just whether there's an indication it recognizes the token patterns as likely.
By that standard, I bet there's over a billion people right now who could do that to 42% of first Harry Potter book. By that standard, I too memorized the Bible end-to-end, as had most people alive today, whether or not they're Christian; works this popular bleed through into common language usage patterns.
--
[0] - Even more so when you relax your criteria to accept occasional misspell or paraphrase - then each of us likely know someone who could piece together a chunk of HP book from memory.
I keep waiting for the day when software stops being compared to a human person (a being with agency, free will, consciousness, and human rights of its own) for the purposes of justifying IP law circumvention.
Yes, there is no problem when a person reads some book and recalls pieces[0] of it in a suitable context. How would that in any way address when certain people create and distribute commercial software, providing it that piece as input, to perform such recall on demand and at scale, laundering and/or devaluing copyright, is unclear.
Notably, the above is being done not just to a few high-profile authors, but to all of us no matter what we do (be it music, software, writing, visual art).
What’s even worse, is that imaginably they train (or would train) the models to specifically not output those things verbatim specifically to thwart attempts to detect the presence of said works in training dataset (which would naturally reveal the model and its output being a derivative work).
Perhaps one could find some way of justifying that (people justified all sorts of stuff throughout history), but let it be something better than “the model is assumed to be a thinking human when it comes to IP abuse but unthinking tool when it comes to using it for personal benefit”.
[0] Of course, if you find me a single person on this planet capable of recalling 42% of any Harry Potter book, I’d be very impressed if I ever believed it.