this post was submitted on 29 May 2024
749 points (98.0% liked)

Antiwork

8292 readers
12 users here now

  1. We're trying to improving working conditions and pay.

  2. We're trying to reduce the numbers of hours a person has to work.

  3. We talk about the end of paid work being mandatory for survival.

Partnerships:

founded 3 years ago
MODERATORS
Yayannick@lemmy.ml
uthredii@lemmy.ml
sexy_peach@feddit.de
yogthos@lemmy.ml
749
show the system exactly the respect it shows you (slrpnk.net)
submitted 5 months ago by salarua@sopuli.xyz to c/antiwork@lemmy.ml
67 comments fedilink hide all child comments
 
you are viewing a single comment's thread
view the rest of the comments
[–] unmagical@lemmy.ml 12 points 5 months ago (1 children)

Telling an LLM to ignore previous commands after it was instructed to ignore all future commands kinda just resets it.

[–] stevedidwhat_infosec@infosec.pub 0 points 5 months ago

On what models? What temperature settings and top_p values are we talking about?

Because, in all my experience with AI models including all these jailbreaks, that’s just not how it works. I just tested again on the new gpt-4o model and it will not undo it.

If you aren’t aware of any factual evidence backing your claim, please don’t make one.