[jerf]

LLMs-as-search-proxies have some pretty nice capabilities. For instance you can say "limit your search to scientific papers" and they'll do a much nicer job. I've also had some success recently prompting them with "I'm looking for reputable sources", e.g. recently I was looking for ways to repel deer from my apple tres. A naive internet search had vendors of shady crap jumping me. The LLMs pulled up relevant papers and university extension programs from my area.

Though I will say I get much better results from the LLMs I pay for than the free ones with Google or DuckDuckGo, which seem to be way way way more prone to just make crap up based on your search and cite web pages that, when followed, don't have the claim being made in the AI search results at all. By contrast every "source" link I've followed in the for-money AIs has 100% backed what the AI said it backed. Don't judge by the free AIs the search engines put out, those things are probably starved of resources and are nearly useless.

(Which I did not intend as a commentary on Google's plans here, but it is a data point of interest... that pressure to cut costs on the "free" services is quite directly at odds with providing quality AI services for the forseeable future.)

[chrismorgan]

Just today, I think that I got a useful citation in DuckDuckGo’s Search Assist (AI stuff) sources for the very first time. The sources it lists have hitherto invariably been already right there in the regular results, or actually not supporting the AI output at all (the far more common case). There was also a useful regular search result in second or third position, but the one in the Search Assist citations was better, and not in the regular search results, even on the second page.

And I’ve tried Google’s once or twice and seen it used once or twice, and used ChatGPT exactly once, last week, and I was not at all impressed by any of them. Their output, for what I’ve personally seen, has been nonsense, obvious, or unverifiable.

[lxgr]

> I get much better results from the LLMs I pay for than the free ones with Google or DuckDuckGo, which seem to be way way way more prone to just make crap up based on your search

Same here. The free version probably gets orders of magnitude less of a compute budget, though, so I am not really surprised.

What I find really surprising though is how many people still have only ever used the free version of any LLM, even those that are heavy users and could easily afford it. It seems like a pretty big and basic product marketing mistake to me to limit capabilities instead of usage time in the free version! How are people supposed to learn what they'd get if they were to pay?