My thoughts on AI overly affirms users asking for personal advice: AI overly affirms users asking for personal advice

Commentary

  • AI overly affirms users asking for personal advice
  • This is partially true, as I said in the above thought to add a “blunt and brutal but honest and grounded” advice, it steers the other way. Its not a sycophancy but more about instruction-following.
  • ChatGPT models are increasingly becoming instruction-following, but are overly sweet sometimes, I agree. Grok is the other way, as we know from the snitch bench.