logoalt Hacker News

theParadox42today at 1:46 AM1 replyview on HN

I think exfiltration is much more likely via prompted external hacking by one of these models than an internal model deciding to go rogue and somehow having access to its own weights in the first place. People do try to exfiltrate model weights indirectly ofc, its called distillation


Replies

skeptic_aitoday at 2:09 AM

Distillation it’s so far away from weights