Getting beyond the title, which I definitely agree with (<a href="https://news.ycombinator.com/item?id=41896346">https://news.ycombinator.com/item?id=41896346</a>), there was this nugget about hallucinations:<p>> I think over the past 18 months, that problem has pretty much been solved – meaning when you talk to a chatbot, a frontier model-based chatbot, you can basically trust the answer<p>Can't decide if he actually believes this, or he's just spewing his own hype. While I definitely agree the best models <i>have</i> reduced hallucinations, going from, say, 3% hallucinations to .7% hallucinations doesn't really improve the situation much for me, because I still need to double check and verify the answers. Plus, I've found that models tend to hallucinate in these "tricky" situations where I'm most likely to want to ask AI in the first place.<p>For example, my taxes were more of a clusterfuck than usual this year, and so I was asking ChatGPT to clarify something for me, which was whether the "ordinary dividends" number reported on your 1040 and 1099s is a <i>superset</i> of "qualified dividends" (that is, whether the qualified dividends number is included in the ordinary dividends number), or if they were independent values. The correct answer is that the ordinary dividends number (3b on the 1040) <i>does</i> include qualified dividends (the 3a number), but ChatGPT originally gave me the wrong answer. Only when I dug further and asked ChatGPT to clarify did I get the typical "My mistake, you're right, it is a superset!" response from ChatGPT.<p>Anybody who says that LLM output doesn't need to be verified is either willfully bullshitting, or they're just not asking questions beyond the basics.