New Delhi, Feb. 7 -- For all the PhD-level intelligence they're supposed to have, our AI assistants can be maddeningly bad at fulfilling requests smoothly. The productive time lost in trying to figure out what is going wrong is so significant that many companies are now evaluating whether the much-touted AI is delivering any real return on investment.

As a user of any large language model (LLM), you'll face annoyances on an almost daily basis. From making things up, to seemingly refusing to obey instructions, to making mistakes that are hard to believe-it can be exhausting. There are serious hallucinations that can cause real damage in high-stakes situations, and sycophancy is another big time waster, other than being dangerous at times....