While I’m glad to see Qwant and Ecosia’s progress, I have to say a search index “designed specifically for LLMs and generative AI agents” was not quite what I was hoping for when I saw this pop up.
Yeah, I was also under the impression that they will create proper search engine for regular users, the AI crap is cancer
I was also under the impression that they will create proper search engine for regular users
It is a proper search engine for regular users
It will power regular search results:
Called European Search Perspective (EUSP), the JV now aims to serve around 50% of French queries and 33% of German queries by the end of the year.
And it also has an API 🫨 that AI search engines can use
It might be a bit naive, but here’s to hoping that it’s just a stepping stone for a proper alternative search engine.
I bet that was the original plan, then they saw that the goal was nowhere near and decided to rebrand a subpar product “for AI”.
How I imagine the conversation went
Qwant, Ecosia:
- we want to build an independent web search index that will greatly benefit European people, European economy, European tech industry and breaks the US monopoly on this area
Government official:
- I have no idea what you are talking about but here is a 200 page grant request that you can fill up and if it is accepted can maybe get 5000€ in 2 years
- We want to build a search index WITH AI !
- OO , I’m calling the president ! you’ll get billions of €.
Yeah I really would prefer we moved away from it… but unfortunately I’m not sure that’s possible. EU can’t afford to ignore AI either.
EU can’t afford to ignore AI either.
Can’t afford to set fire to large piles of money?
I’m being slightly flippant but I don’t really understand what you meant by that.
EU can’t be in a technological disadvantage when it comes to AI. Of course it wants to make it possible for European AI devs to have support, like a search index that also conveniently has the potential of favoring information that aligns with EU interests.
EU can’t be in a technological disadvantage when it comes to AI.
I think we might be at an impasse as “AI” currently seems to be a solution looking for a problem as far as I can see. Not that LLMs are completely worthless but the “hoover up the internet and hope something useful comes from it” approach seems to be a bit of a technological cul-de-sac that I think the EU would be better off ignoring. Let the other nations spaff money and resources on it, learn from their mistakes, then make a move. We’re long past first-mover advantage after all, so we’d only be playing catch-up. Why not just wait a bit and leapfrog?
You could be very right but I doubt powers that be want to take that chance. That said, I do consider AI dangerous in the world of information warfare.
Not that LLMs are completely worthless but the “hoover up the internet and hope something useful comes from it”
You’re saying that currently AI does not help with anything?
I’ve not found them reliably useful in a personal or professional capacity for anything other than trivial tasks. I’ve spent a lot of time reviewing code written by LLMs and it was not a good use of my time.
A technology that is inherently unreliable and requires constant supervision needs to be on the level of nuclear fission to be worth the hassle, in my view.
Note that I specifically am talking about “hoover up the internet” LLMs, not specialist tools built on scientific datasets or whatever.
It can summarise my email? Fantastic, except it cannot be relied on to do a good job so I might as well do the reading myself as at least that way the data I’m getting is correct.
deleted by creator PIEFED
search has been using ai under the hood for many years, long before llms came along. that’s why google has had so much ai research, even back in like 2012. it’s one of the best ways to keep track of that much data.
like it or not the field is very useful, llms are just a rather stupid user-facing part.
Hey if they’re actually useful then that’s fine! The useful stuff hasn’t really been what the AI bros have been pushing in recent years.
no, that’s fair. we usually don’t see the useful stuff because it doesn’t need hype and just shows up as a data point.
And this is why Lemmy is better than Reddit. Not everything has to be a knife fight 😆
as long as we’re not in a knife fighting kinda community, or i manage to make the mistake of criticising north korea again… then yeah
The anagram isn’t that hard to find, not a great sign they overlooked that…
I don’t think anyone really cares about things like that in 2025
