Hacker News new | ask | show | jobs
by thfuran 713 days ago
Even if that's true, the example definitely isn't proof of that given that the old technology seems to handle the problem just fine.
1 comments

If they each give the correct answer in the same number of steps then there’s no reason to prefer one over the other.
My multiple experiences with ChatGPT hallucination, even at minimum temperature, leaves me preferring a web search answer. Just last week, jq hallucinated three different jq syntaxes that failed to parse.

That being said, I had been using web search for several years looking for the Get Smart episode I had seen as a kid in the 1980s where the enemy agent explains that he can't give up his suicide ring because "That's how it works, if I take it off, my wife will kill me." Google Bard found the episode on the first try.

So, I tend to fall back to LLMs only when search engines fail me.

Mental inertia is the phrase I use for preferring the one you learned first
Power usage, latency, ethical concerns.