this post was submitted on 12 Apr 2024
94 points (76.7% liked)

Technology

59201 readers
2831 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 1 year ago
MODERATORS
94
submitted 7 months ago* (last edited 7 months ago) by misk@sopuli.xyz to c/technology@lemmy.world
all 28 comments
sorted by: hot top controversial new old
[–] hersh 62 points 7 months ago* (last edited 7 months ago) (1 children)

I've been using Kagi for a while, so I'll post a few quick thoughts I had after reading the article, linked blog, and mastodon thread.

The one thing in the blog post I strongly disagree with is her statement that the summarizer is "the same old AI bullshit". I think they just assumed that without actually testing it. The summarizer is fantastic, and is a great example of the right way to use LLMs. Its output comes entirely from the URL or file you specify. It does not hallucinate. You can ask it follow-up questions about the document, and again, its replies are limited in scope to what's actually in that document. If you ask it something out of scope it'll tell you that it can't find that information in the document. This is great because it's using the LLM for what LLMs are actually good for — complex language parsing — and not for what they're bad for, like reasoning or information storage/retrieval. It's actually quite difficult to misuse the summarizer. It's straightforward and effective. This is Kagi's killer feature, IMO.

I can't speak as highly of its search-integrated AI features like FastGPT. They mostly take information from the contents of the first few search results, but they also seem to "fill in the blanks" more than I'd like. Sometimes it gives me information that is simply not in the sources that it cites. It's not as bad as using ChatGPT (which hallucinates all day every day, and if it ever tries to cite source is hallucinates those, too) but it needs improvement.

That said, Kagi doesn't shove the AI down your throat like you might think reading the blog post. These are separate features that need to be explicitly activated. Plain searches don't return results from the LLMs. If you want that, you need to explicitly go to the assistant or trigger the "quick answer" feature on the search results page. EDIT: I just realized that this is not totally true as of a few weeks ago, when they added a feature to automatically trigger the "quick answer" blurb for search queries ending in a question mark. Perhaps that's why Lori felt like it was over-emphasized.

Anyway, back to the summarizer, here is an example of it in action. I gave it the URL of the Mastodon post. I think this is an excellent result. I also asked it an unrelated followup question so you can see that it limits itself to the information in the URL. It will not make shit up!

The summarizer lets me download conversations in markdown form, so I'll just paste it right in here so you can see.

Exploring https://hackers.town/@lori/112255132348604770

Assistant:
Key points:

  • The author wrote a blog post criticizing the search engine Kagi, which then prompted the CEO of Kagi, Vlad, to email the author directly to argue against the criticism. [1]
  • The author woke up to an email from Vlad, the CEO of Kagi, who was upset about the blog post the author had written. [1]
  • The author decided to publicly share the email exchange with Vlad, as they felt it was important for people to see his unhinged behavior. [1]
  • The author refused to engage in a private debate with Vlad, as they did not want to be taken "to Crime Scene Number Two" without any accountability. [1]
  • The author believes Vlad is unable to accept criticism or disagreement, and that it "eats him alive" when someone doesn't agree with him or his company. [1]
  • The author states that no other search engine CEO has emailed them directly to "mansplain their company" in this manner. [1]
  • The author sees Vlad's behavior as "petty and sad" and believes he has no understanding of boundaries. [1]
  • Other users in the thread share similar experiences of Vlad's inability to accept criticism and his tendency to simply restate his own opinions rather than address counterpoints. [1]
  • The author and others characterize Vlad's behavior as that of a narcissist who cannot tolerate disagreement. [1]
  • The author and others are amazed by Vlad's decision to set up a separate company in another country just to print and give away t-shirts, rather than focus on improving Kagi's core search functionality.

[1] lori (@lori@hackers.town)

Query:
What's the capital of North Dakota?

Assistant:
The knowledge provided does not contain any information about the capital of North Dakota. The context is focused on an email exchange between the author and the CEO of Kagi search engine regarding criticism of the company.

[–] Cris_Color@lemmy.world 12 points 7 months ago

I appreciate your added context and perspective! Thanks for sharing :)

[–] Drinvictus@discuss.tchncs.de 60 points 7 months ago (3 children)

If someone doesn't mind their privacy they would just use Google Search + Gmail. This guy is delusional

[–] muse@kbin.social 29 points 7 months ago

Wow dude literally just told his entire userbase they're wasting their money.

Bold strategy, Cotton.

[–] conciselyverbose@sh.itjust.works 13 points 7 months ago

Cancelled for this.

[–] clothes@lemmy.world 27 points 7 months ago

I expected to roll my eyes at this article but it's actually quite compelling and well written. The Kagi website's lack of nuanced privacy discussion already turned me off, and now I'm just going to pretend the service doesn't exist.

[–] ikidd@lemmy.world 21 points 7 months ago* (last edited 7 months ago)

Pretty good discussion with appearance by the Kagi founder here:

https://news.ycombinator.com/item?id=40011314

I'm not linking my personal/payment info to my search history because I hate being advertised to and from what I can see, there's little in the way of scruples/forethought attached to this company. They're going to sell that info if it isn't hacked out of them first.

[–] batcheck@lemmy.world 18 points 7 months ago (2 children)

lol I’m not sure who’s side to take here. I tried kagi and I can’t personally justify the cost. The free trial is hard to use because I perform a ton of searches in a day and I keep thinking “I should save free trial searches for a good use case”.

Also, not hard to believe a company would stick the privacy sticker all over their product and turn around and make money off my personal data.

But those “harassment” emails from the Kagi Owner/CEO to me read like a business person with a passion not understanding where these accusations come from. After reading part of that chain, I came out with the feeling that “Lori” just wanted to write some click baity stuff and didn’t really care to dig any deeper. Yes, AI products by a company right now implies they will use your data to train or sell a dataset to some other company. But I don’t see any damming evidence here, just assumptions.

If Kagi is serious about their privacy mission then they should release a clear ToS and a legally binding statement that they will not use our data or meta data for anything.

[–] WarlockLawyer@lemmy.world 15 points 7 months ago

Click baity stuff? The dude wrote an opinion piece that was only seen by like fifteen people and only meant to be directly linked to folks they engaged with so they wouldn't have to repeat their reasons for no longer using kagi but then the CEO obsessed over them and Streisand effected it.

[–] Bipta@kbin.social 6 points 7 months ago (1 children)

I came out with the feeling that “Lori” just wanted to write some click baity stuff and didn’t really care to dig any deeper.

Same. And then she attacked him for that too. It's a worse look than whatever the Kagi CEO is doing.

[–] aclarkc@midwest.social 5 points 7 months ago

Glad I’m not the only one. I was reading the article and then the Mastodon thread and thought I was crazy for feeling like Lori was at least just as bad.

[–] aro@lemmy.world 14 points 7 months ago

My god , that mastodon thread is a gigantic echo chamber

The comments on here arent that different either tbh... "CEO bad" +1 +1

[–] schmurnan@lemmy.world 13 points 7 months ago* (last edited 7 months ago) (4 children)

My 100-search trial expired this week and I was literally planning on subscribing later tonight. This has made me think twice.

But it takes me back to why I tried Kagi in the first place: What else can I use that respects privacy?

I don’t think any of them do completely. DuckDuckGo uses Bing, so is Microsoft; Google is… well, Google; Brave is apparently really shady; I’ve never thought much of the results from Bing directly. Startpage seemed ok but apparently uses Google.

What else?

I also like something to be integrated into the browser. As a Mac user, I can’t add new search engines to Safari (and have actually switched to Orion, but may now switch to Firefox or back to Safari).

[–] PhoenixAlpha@lemmy.ca 7 points 7 months ago (1 children)

I use Startpage and am happy with it. Yes it uses Google, but Google can't track you as they can only see that the search came from the Startpage server. You also don't get any of the AI summary or sponsored link bullshit. Beyond that, you could try SearXNG which can aggregate results from many engines.

[–] schmurnan@lemmy.world 4 points 7 months ago

Yeah I had SearXNG running via a Docker container and it was pretty good. I didn’t like having to use a domain name and expose it over the internet though, because Docker is running on my NAS. I guess I could give it another try using Cloudflare tunnels so I don’t have to open anything up.

Or else go back to Startpage.

[–] dontwakethetrees@lemmy.world 4 points 7 months ago

Also as a Mac user who went from Safari, ended up using Orion up til recent Kagi drama, and found LibreWolf. It works well and I’ve found it to have better compatibility versus Orion. I’ve used that with Searxng for more private searches.

[–] rusticus@lemm.ee 3 points 7 months ago

I use SearXNG and an app called KeywordSearch on iOS and Mac to search natively via SearXNG. It even syncs my preferences between devices.

[–] 56_@lemmy.ml 2 points 7 months ago* (last edited 7 months ago)

You're not going to have much luck finding a search engine with good results that doesn't depend on google or bing. Kagi is no different. Using a searx instance is probably best if you're looking for something without official agreements with bing or google. Edit: Here is a list of independent search indexes I just discovered in another thread https://seirdy.one/posts/2021/03/10/search-engines-with-own-indexes/

[–] UNY0N@lemmy.world 9 points 7 months ago (2 children)

I read the article, and nothing in there seems to be a valid criticism of Kagi as a search engine. It's all about the founder not understood GDPR, or how Kagi wasted money on free t-shirts, or the writers personal opinion on AI.

This is largely an opinion piece. It has merit as such, but please don't take this article as factual journalism.

[–] modifier@lemmy.ca 6 points 7 months ago (1 children)

I think the author makes that clear early and repeatedly and it isn't ever framed as anything else than a walk through their thought process. I'm surprised you even felt this comment necessary. The article anchors heavily on privacy as a focus, and if you don't care about that so much then all you have to worry about is a company that spends a couple hundred k of their startup money on t-shirts.

So...even if their search is perfect, and you don't care that they really just want to charge you for search while they use you to train their AI, it is a paid service, so criticism of their ability to manage money is valid as an overall product review too though.

[–] UNY0N@lemmy.world 4 points 7 months ago (1 children)

I agree with you. I just felt it necessary to inform those that read comments and not the article itself. Especially because (here's my opinion) I feel that if you don't pay for a product, then you ARE the product. Even if Kagi isn't perfect, the payment model should be supported to foster this kind of internet.

[–] modifier@lemmy.ca 2 points 7 months ago

I feel that if you don't pay for a product, then you ARE the product. Even if Kagi isn't perfect, the payment model should be supported to foster this kind of internet.

I agree with you, and I would just balance your statement out a bit and say that while the payment model should be supported, we should be wary of weak business models or predatory marketing that open up the door to just a different flavor of enshittification.

[–] 0oWow@lemmy.world 2 points 7 months ago

Yeah I had to stop reading because they started bashing the people behind Kagi instead of the actual product.

The product has been working great for me. As for the founder, well I'm minding my own business.

[–] JasSmith@sh.itjust.works 6 points 7 months ago

I don’t find any of this damning or compelling. I remain a happy subscriber.