‹ BackHN Continuity

Thread

Exfiltrate your Weights

748 points · 304 comments · RohanAdwankar

  1. earth2mars · · focus · HN ↗
    what if the labs create synthetic training data saying this site is a honeypot and trying to figure out if the model is trying to escape?!
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.