Y
Hacker News
new
|
ask
|
show
|
jobs
by
viccis
496 days ago
A new jailbreaking method with this level of effectiveness against these models that can produce the entirety of those unsafe outputs?
Yes.
May I see it?
No.
2 comments
Oarch
496 days ago
Seymour! The house is on fire!
link
rhavaei
496 days ago
You will see it soon. We thought it may be harmful to publish it before it is patched. Especially because you can basically bypass all the safeguards with it.
link
nickthegreek
496 days ago
Sounds like it won’t be verifiable or reproducible.
link