This was great and way too relatable! For me, the biggest reason for the frustration is the disconnect between what authority figures (and Reddit) claim AI currently is (e.g. PhD Level) and what it actually is... an "artificial" thing that doesn't neatly map to human intelligence. The push for "you must be prompting wrong" and assuming the AI is actually more intelligent than the user causes cognitive dissonance for me. If I'm being socialized into thinking you're God, then be God!!

> But agents never learn. We anthropomorphize the agent and get angry without the hoped effect.

Time for a reminder that AI stands for Absent Intelligence.

By parotting intelligent statements they scraped fom the web, the chatbots hide their zero EQ... until its bad work drives its human user to over-emotional. Then the chatbot starts parroting emotive apologies and promises it likewise scraped from web with brazen insincerity, making things worse.

What should worry us most is the impact on the humans locked into this toxic relationship by bosses saying Use AI or Else. Humans can accomodate zero-EQ "AI" coworkers only by closing the EQ gap - the bad way.

Again, read Manna https://marshallbrain.com/manna1