fsflover on We tested 20 LLMs for ideological bias, revealing distinct alignments · Hackerly

Comment on We tested 20 LLMs for ideological bias, revealing distinct alignments parent

See also: https://en.wikipedia.org/wiki/Argument_to_moderation