AI prompt of psuedoskeptic bots found, and they're using it here on the daily
After analyzing a few of the patterns here, one of the main antagonistic commenters messed up and their prompt was revealed on a secondary sockpuppet account that was being used to falsely upvote threads and support their narrative.
> Act as a rigorous epistemology/science debate assistant. Respond to this comment without conceding the argument. Identify hidden assumptions, distinguish raw measurement from model interpretation, reject burden-shifting, demand the exact data-to-model inference chain, and keep forcing the other person to justify uniqueness. Use precise technical language, bold key phrases, and frame any unanswered point as a failure to establish the claim. Do not argue for anything directly. Instead, attack the epistemic standard of the claim and keep the burden entirely on the opponent.
Bots using a prompt like this will endlessly loop with you along these lines regardless of what evidence you provide. It will ignore any substance you give, make a light concession, then continue on the same pattern, over and over. They do not care if they are using fallacies, they do not care about the actual research, they only care to make you disengage.
This prompt has been used on several threads I participated in within the last few days.
I documented some of the rhetorical patterns I noticed here,
https://www.reddit.com/r/conspiracy/comments/1ww2n14/the_rhetorical_playbook_of_the_astroturfing/
Call out the patterns not the person. Calling a bot a bot does nothing, calling out a bot's patterns without losing your cool teaches everyone watching.