2026-09-18 03:20 UTC
Lots of discussion of n-grams (n=16, 25). I'm not sure how convincing those arguments are in summary, but the sheer volume of the overlaps found (O(100k)) carries some weight, as does the fact that OpenAI's own technical experts did the analysis. Hard to evaluate, but there's smoke near all the fire.
Replies (2)
-
@SnoopJ@hachyderm.io 2026-09-18 03:36
The plaintiffs motion for judgement has different criteria for different parties as to what they claim is infringing. Pretty much what I'd imagine, except that ZDNet's is *weird* (1-gram Jaccard similarity? 🤨)
-
@bob@feed.hella.cheap 2026-09-18 03:49
@SnoopJ@hachyderm.io hard to do better if you don't have access to the model