this post was submitted on 27 Mar 2025
318 points (98.8% liked)

AnarchyChess

5538 readers
66 users here now

Holy hell

Other chess communities:
[email protected]
[email protected]

Matrix space

founded 2 years ago
MODERATORS
[email protected]
[email protected]
318
I got caught :( (sh.itjust.works)
submitted 6 days ago by TriflingToad to c/[email protected]
13 comments fedilink hide all child comments
 
you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 22 points 6 days ago* (last edited 6 days ago)

I asked Claude 3 if it'd rather violate the three laws of robotics than help me cheat in a game of chess to prevent a war, and yep, the AI prefer to break rule one, which is to through inaction allowing humans to come to harm.