I agree that Google is terrible at its job and produces garbage most of the time. That's the only reason why an LLM chatbot is seen by anyone as a viable replacement.
But at the very least Google always gives you the source, which allows you to read it for yourself and judge its trustworthiness. No such option for ChatGPT.
I can't count the number of times I've clicked through to a link expecting to find more context than the excerpt on the search page and that content isn't there. Instead I get trapped in a sea of circular SEO spam that all promises to give me the answer I originally searched for but ends up leading me nowhere except to other article summaries with content I didn't even want to read about. Where's the source?
A good chunk of the remainder of the times I click through - especially on work-related stuff - it turns out the page is some kind of cribbed content from GitHub, StackOverflow and so forth, but mangled together and reskinned with different randomly-generated usernames making the "comments", so it seems like it's a different site, and you can only really tell by picking a specific sentence that appears unique and then manually searching for that same sentence on several different sites. No clear source there either.
Web search is broken, and it has been for several years now.
> I can't count the number of times I've clicked through to a link expecting to find more context than the excerpt on the search page and that content isn't there.
Excellent: this quite quickly tells you that the search result was wrong. So you quickly got very relevant context and can tell at a glance that there is no source. You don't get any such context when GPT provides an invalid answer.
> Google is terrible at its job and produces garbage most of the time
Are we using the same search engine? I use DuckDuckGo most of the time but Google is hardly terrible, even if it could be improved.
I find what I’m looking for in the top results 90-95% of the time. The other 5-10% also contains times where there simply isn’t a good link that solves what I’m looking for even after digging hard.
DDG is much more straight forward vs Google often tries too hard and fails.
It's like how Twitter's default feed changed to try to tell you what you wanted to see, while the old chronological feed was the content you originally asked to follow. The simpler feed was always better and less contentious for non-newbie/casual users.
Google is better at deciphering what you want, but DDG is straight up: here's the simplest results based on your query.
Maybe the solution is both? An evolution of the ML approach that Google uses (ala ChatGPT style mimicking a human) AND offering a tab, or box, or whatever interface, with 'simple' straightforward old-school search results. Until the former is better 99% of the time, they shouldn't abandon the latter.
I was pleased at first when I tried to use ChatGPT as a search engine. I asked something like "give me the top 5 blog posts to learn about C++ metaprogramming" and it actually spit out extremely plausible titles from top people (e.g. Dave Abrahams, Eric Niebler, Herb Sutter, etc), but when I said "no please give me the actual URLs", they were all 404.
If Microsoft somehow is able to implement the information synthesis (e.g. "give me an example of C++ metaprogramming, using Boost Hana") with actual indexed links from sources, that would probably be the one thing I'd see in the near-ish term that could upset the search engine balance of power.
I don't know if this is feasible or not, but having the chat bot also be able to respond with a subset of sources that it based its answer on would be a satisfying combination of both worlds I think.
I would be really curious to see how trying to learn a new subject via a chatbot that is able to also provide source material links would go.
> I don’t know if this is feasible or not, but having the chat bot also be able to respond with a subset of sources that it based its answer on would be a satisfying combination of both worlds I think.
But an LLM doesn’t look at particular sources, find relevant ones, and write a response based on them, it is a mathematical summation of the way all of its sources look as sequences of text, and produces a response that based on that fits with its prompt. Identifying which sources most influence a response isn’t merely something that they don’t currently do, its information that, AFAIK, there is no plausible way to recover from the model.
After I commented I did do a little bit of thinking about the little that I know about this subject and realized that it probably isn't feasible to extract that data. Thanks for the comment because... you are right as far as I know. It is a bit disappointing to realize this though.
For most programming questions at least, the highest placed result links are StackOverflow answers related to your query (if any exist, of course). I would say this alone invalidates your "virtually guaranteed" claim.
But at the very least Google always gives you the source, which allows you to read it for yourself and judge its trustworthiness. No such option for ChatGPT.