Thread

1. theParadox42 · 1789868810 · focus · 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

   1. skeptic_ai · 1789870165 · focus · HN ↗

      Distillation it’s so far away from weights

Unofficial Hacker News client; not affiliated with Y Combinator.