Anyone else notice that LLM's are overwhelmingly referencing Reddit threads to source answers? The internet must be seeing this as an opportunity to spike AI answers by low-key injecting fraudulent information. Just like left-progressives are doing in their attempts to rewrite science, history and all the rest.

I wonder if they've excluded certain subs from inclusion in the training data for LLM's.