Modern AI chatbots have a serious problem. During the fine tuning process, data scientists aren't very careful and don't involve disparate perspectives or more than one human speaker. As a result, AI conversations rarely confront humans with new perspective, and have a very concerning tendency to devolve into shallow flattery and repeating your own beliefs back at you.

Because of this, I highly recommend for any AI conversation you have, that you refuse to read or engage with any AI statements that rephrase something you said, and any AI statements that aren't answering a non-rhetorical question you asked. I recommend only reading and responding to AI statements that directly contradict something you said, answer a non-rhetorical question, or ask a question that you haven't already answered. #AI