- cross-posted to:
- [email protected]
- cross-posted to:
- [email protected]
science shows as true what you thought was only 99% true
https://www.youtube.com/watch?v=uVf7VUX_iUk&list=UU9rJrMVgcXTfa8xuMnbhAEA - video
https://pivottoai.libsyn.com/20251015-ai-is-not-popular-and-ai-users-are-unpleasant-asshats - podcast
time: 5 min 57 sec
The only thing that drives me to AI is the extreme uselessness of modern search engines. This is not an endorsement of hallucination engines as much as it is a condemnation of late stage enshittification of search engines and the internet in general. I miss the days when I could google something and actually find what I was looking for.
… the search engines became crap because of ai
Plus, ai just lies. It’s not a replacement
Shitty search has been a problem longer than ai
If your definition of ai is only LLMs, sure. But it’s been algorithmic tweaking and SEO wars for a while now.
Well yeah, llm garbage is the problem being discussed. Do you consider SEO to be ai? I don’t see why you would.
No, but the algorithms that search engines use to combat it is.
ah minor correction, you’re on Lemmy not on Mastodon, the venue for this sort of tedious and point-avoiding pedantry
No, they aren’t.
AI has been around since the 50s, Internet search has only been around since the 80s.
That is irrelevant. We are specifically talking about LLMs ruining Google searches, which is the last couple of years.
I mean, search has always been built on some kind of LLM. That’s how you convert a query into a list of page-results.
We’ve just started trying to wrap the outputs into a natural language response pattern in order to make them feel more definitive and more engaging. The “AI” part of search is mostly window-dressing.
It has inaccurate heuristics and often tries to back-fill what it can’t find with an approximation in order to maintain user engagement.
Idk if I’d even call it lying, so much as bullshitting.
no it fucking hasn’t. the stemming and page ranking algorithms used in traditional search have absolutely nothing to do with LLMs.
shit, neither stemming nor PageRank as originally defined even have a machine learning component. here’s postgres’ full text search suite, which literally converts a textual query into a list of results (sans page ranking, which is out of scope for a database) in a manner suitable for a production search engine, utterly without any machine learning or other stochastic crap.
It’s not perfect, but https://udm14.com/ is at least an improvement. I found it too tempting to read the AI answer that pops up right away and is often wrong.