The negative feedback is curbing my responses.
Almost like I am a neural-net (wet one), and I am receiving RLHF (downvoted), which is modifying my internal network weights on future responses.