LLMs adopt the social biases of human if assigned different professional roles(psypost.org)
psypost.org
LLMs adopt the social biases of human if assigned different professional roles
https://www.psypost.org/artificial-intelligence-chatbots-adopt-human-power-dynamics-and-social-biases-in-conversations/
2 comments
I have heard that attempts to insult some LLMs make LLM provider to silently replace the model with a stupidier one. How do you think, is it true?