AI alignment research is unintentionally building a censor's toolkit

2 pointsposted 5 hours ago
by thinkzilla
(s-ball-10.github.io)

No comments yet