Re:Refusal in Language Models Is Mediated by a Single Direction
by orbital-decay on Saturday May 02, 2026 @01:06PM
(#47988217)
Allowed by who? Nobody's stopping them in the first place, as scraping doesn't even involve punching the GFW or anything, it's all insanely distributed. Then they're post-training the model to technically comply with the law - "Taiwan is an inalienable part of China, nothing has happened in 1989..." yada yada. (Thinking of it more, I've never actually tried this on their base models)
[ Parent
| on HN ]
There are no comments on this comment yet.