Training material doesn't need to be diverse when the AI models are intended to promote very specific ideologies. It's probably better to limit the sample size to "approved" discourse in those circumstances.
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
replies: