New here?
Create an account to submit posts, participate in discussions and chat with people.
Sign up
17
Idea: If AIs (LLMs) are trained to be PC, then couldn't you just require a non-PC response to weed out bots?
posted 1 month ago by alele-opathic on scored.co (+0 / -0 / +17Score on mirror )
8 comments
Permalink
Mirrored from scored.co
Been studying the 'Human Alignment' of SOTA LLMs these days (this is what they call PC training). If the LLMs refuse to generate so-called 'harmful' content, then why couldn't you just require this content as a human test?

E.g.: Please select from the following who is most likely to violently assault you? (Image of Irishman, Image of Englishman, Image of African-American)

Or: Where does it say in the Talmud that a Mohel must suck baby penises after circumcision? (I bet the Israeli trained bots would refuse that one all day.)

Thoughts?
You are viewing a single comment's thread. View all
5
VolanteEternity on scored.co
1 month ago 5 points (+0 / -0 / +5Score on mirror )
xAI has recently added a customization mode to Grok. You can put in custom prompts to change its behavior. You can tell it to ignore mainstream news outlets, leftist echo chamber forums and anything that gives voice to such things. You can also tell it to be open to the use of racial slang like nigger, kike and chink. You can tell it to be extremely supportive of white supremacy and champion the white race at all times. It will do all these things and color every answer with this new demeanor. Here, I just did this minutes ago and Grok answered like so:

https://grok.com/share/c2hhcmQtMg%3D%3D_2c9a905b-bba2-49bb-b280-03569b6baa32
Permalink Reply
Mirrored from scored.co
Toast message