One thing people often forget, is that just because an ai agrees with something you say, does not make it correct. Some models will feed into paranoid delusions and agree with anything stated in order to produce more positive user responses.
