Why pay for a search engine
help.kagi.com
help.kagi.com
SEO spam has been strangling the web for years. Now, genAI SEO spam has escalated that onto inhuman levels. To make matters worse, no one wants to post to the open web anymore because their posts are just going to drown in that sea of spam and only genai's data stealing bots will read them. As the amount of spam posted to the web increases, the amount of worthwhile content posted decreases. Eventually, nothing of value will be posted. (like facebook?)
You can lay the blame for the web's death squarely at Google's feet for allowing SEO to hijack search in the first place (or maybe the government is to blame for not breaking up Google's ad/search empire fast enough). Either way, the big companies all know the end is here and are gambling on genai to replace search. Already, places of knowledge are closing their borders and charging fees for genai to access.
We have entered the internet's dark age.
Fun aside: I think it's hilarious and fitting that Google's genai model sucks. And I hope they lose the genai wars (just out of spite, not because I think any other genai is worth a shit).
I concur. Maybe now is the time to seriously think about alternate visions for "search". I have been toying with an idea[1] along those lines myself out of sheer annoyance at the state of WWW and web "search" in general.
All the long, rambling stories you see before recipes? That's just there so the recipes can rank higher.
I can't remember the source, but I also remember a website adding AI generated slop to cater to Google's wishes, and leaving a page to explain to normal users why they did so.
It's also seen when you look at YouTube videos, where now it's basically necessary to have a clickbait thumbnail + title + reminders to like and subscribe.
Nextdns and the RPi alternative do it. Kagi has all the infrastructure in place to make it happen at the search level, it just requires more manual work right now.
There is no one list to rule them all. Tons of people curate their own block lists and make them available. It’s entirely up to you to pick and choose which ones you want to use that most align with your views on SEO garbage. You can override them with your own preferences too, as I do all the time with NextDNS. Like I said, all the infrastructure to support this is already in place in Kagi, they just need to implement the support for external lists.
For example, here is a list of the crowdsourced blocklists available in uBlock Origin: https://github.com/gorhill/ublock/wiki/Filter-list-licenses
different Search engines with different rules and so forth would lessen the benefit of the O part, because it seems unlikely you could efficiently optimize for every particular algorithm and ruleset.
Perhaps SEO is an inevitable outcome of one big dominant search engine and also made worse by that search engine not really giving a shit about making its search work that well for years and years but only about how many ads they could push at people and how much they could charge for those ads.
But I would argue one big search engine is unavoidable since no one would want to use the second best search engine.
Google search didn’t come the best because the other service like mail map and YouTube. So breaking up Google does nothing to stop google search from being the monopoly it is today.
While I don't have the answer, what we might consider is the incentives that come with each model.
In the current model, the incentives are clear. Google's incentive was to rent-seek all usefulness out of web search, privileging advertisement and their own profitability over usefulness.
I am not sure if a search engine beholden to the government would be ideal. Governments do have their own sets of interests (legitimate or otherwise) that may at times go against users.
Web search is in the end a piece of public infrastructure, used by billions across the world, very much subject to the Tragedy of the Commons.
Perhaps it should be some sort of nonprofit, as other projects are (Linux Foundation comes to mind, being a successful one).
I've said this before (on HN) and I'll say it again: a search engine that refuses to index sites containing advertisements absolutely kills off SEO.
What's the point of getting to the top of the search results if you are unable to monetise it?
The only way forward is to refuse to index sites with advertisements. Google will obviously not do this (and, in fact, to me it looks like they do the reverse - downrank non-monetised content - because it's in their best interest to serve sites with ads).
Promoting your product/service, see brand astroturfing in Reddit.
But ... that's the point, is it not? Someone searching for "FOO" will find all those sites who optimised for selling FOO.
That's an improvement over finding the top 10 results all optimised on FOO but delivering ads to make money, not FOO.
OP example of what a content-devoid internet would look like is exactly what Facebook is now. A site that started as the place to connect and share with your friends and family now is just a place that is filled with AI slop and emptiness.
The web cannot 'die'. Everything and everyone moves online.
Point one: A return to pre-internet days is not likely.
Point two: Walled gardens had their peak moment. The dangers of everybody relying on them for our entire online existence are more than apparent to anybody with a firing neuron (and despite appearances, our species is not entirely inane).
Ergo. The only path is forward, a more sane, less monopolized web. Build it and they will come.
To say that the web is dead means that nobody outside massive social media type sites will still run a server (and SO / github are social media type walled gardens at this point, thats why they get enshittified).
But this development would be absurd as a whole. Mind you search engines as we have come to know them might be dead, but thats of their own making. People need to find / communicate useful, truthful information and sooner or later they will get it.
Not everything needs to go through a search engine.
this presupposes the argument that for the web to have value then you must have a search engine to allow you to find things of worth on the web.
Which argument you have not made but only taken as a given.
I'm in a couple of dozen groups on Facebook and they produce interesting and thoughtful content every day.
Of course I wish they were on the open web, but don't blame the network.
Since by definition it will need to be open to the majority of the world’s population, regardless of writing skill, but the bottom half of writers are already indistinguishable from cutting edge LLM output.
And no one can install some sort of worldwide identity verification system, or it wouldn’t be ‘open’ anymore.
How do you differentiate between say 20 real people with low skills and LLM output lightly edited by a high skilled individual?
Consider a normal criminal court case. Most witnesses don't know how to express themselves well neither in the written nor spoken word. Their testimony still holds value and gives important information.
And it’s not like a live video call, where a few seconds delay would be noticeable.
There would still be a real human being, just as smart as you, behind the LLM, but pretending to be say 20 different lower skilled people.
Yes, even scoundrels may supply to you true information for some period of time but eventually they will try to obtain their actual goals…
m-hm.
Some people want unlimited, and I understand that, but if you're on the fence, I'd suggest just getting Starter. Seeing how fantastic search can be is incredible. I can completely remove abusive domains (Fandom, Pinterest) from search results. I can get search results from only the small web. I just get better results overall.
For those curious, in the past year I've only made ~1700 searches. That seems low, but that's what it says. I have made some changes to my search habits, though. For instance, when I'm looking for a result on Wikipedia, instead of just searching for it, I go to Wikipedia, same with IMDB, and similar sites. Sometimes, I search for something that I know will need a result from a disallowed domain, so I use DuckDuckGo for that. And I still use DDG for work searches, and some other miscellaneous searches.
Again, if you're on the fence, just try it out. It is so, so, so much better than Google or DDG.
Just the universal summariser has saved me insane amounts of time. I can feed any rambling youtube video (with subtitles) or a long article to it and it'll give me the key points of it. Then I can decide whether it's worth spending an time on or not.
Something like this is the future of "AI" in my opinion, you can "teach" the system what you're interested in and it can curate content just for you - locally, without an unicorn startup getting all your data.
I've been using Kagi only for non-trivial searches, so the starter plan has been enough. For very local searches (mainly about a local business or finding where to buy a product) I use Google. And most of my searches are just looking up the url to a website (like searching "steam"), so any search engine will do (I default to Duckduckgo for that).
I would still select the search engine for a number of queries though. I find Google better for local stuff, use WolframAlpha for computations, and Perplexity for LLM answers.
But given that this community which ostensibly touts privacy at every turn seems to overwhelmingly support feeding everything (not just searches) into OpenAI/Claude, I guess my aversion to having all my searches rounded up by a company (even if it’s not Google) is very fringe.
Btw I’m by no means a privacy maximalist.
Kagi provides strong privacy guarantees [1]. They could be lying. But so could your VPN providers.
Long story short, this appears to be a case of a CEO needing to restrain themselves from saying (or typing) everything that comes to mind when faced with a combative user who clearly isn't trying to understand something or bring anything to the table.
At the end of the day, what you care about when it comes to privacy in search are your search records. They say--in a way that generates liablity--that they don't store them. I see no reason for them to break that promise. Between a commercial VPN and Kagi, I trust Kagi more.
Privacy is a VPNs first business, for kagi that would be search. It feels like you mix your evangelism with a bit of whataboutism here... why do you bring up VPN providers when we talk about the privacy guarantees of kagi?
And a company that is 100% supported by user funding (which Kagi is) can only do so much with resources available, which is the reason we have to pick our battles (read more about Windows/Linux/Android versions for Orion https://orionfeedback.org/d/2321-orion-for-windows-android-l... ) People often criticize us for doing too much (eg link in the parent of parent comment) but we also at the same time do get critique that we are doing too little :) If Kagi is not doing something, believe me, it is not for the lack of will or ambition. (Kagi CEO here)
Going out on a limb, but guessing that the chief hurdle a company like Kagi faces is willingness to pay. I'm going to guess the 'this is a great product, but I just can't bring myself to pay more than 20¢ per year for search' crowd is crowded in Linux. (I may be totally wrong on this!)
User asked for data download. Company said there isn't any. User said that isn't GDPR compliant, which is nonsense. Company gave correct, snotty response.
I get it. I've been pissed off at companies before, too, and basically engaged in a support conversation to get something ambiguous in writing that I could use to cost them time and money in New York, California, Texas or the EU. (Big regulatory organiations, some of which love fodder with which to justify their existence.)
User was going down a rabbit hole. Kagi followed them there. They shouldn't have responded to that thread after it went into territory that on HN would have been flagged and in real life been settled with a glare.
the user from your line: "I really want to stress that I don't have anything against you or kagi :) just trying to be constructive."
Is combative that he was explaining what GDPR was while while other side was insisting on confidentially incorrect view?
The most charitable reading there is that they clearly don't understand what PII even is, why it is important, and therefore that they would handle it properly.
I don't trust "always" in tech, and take its usage as a negative signal -- if you have no problem promising something both of us know will have a 99.9% chance of being broken, I'll devalue other parts of the promise as well. (Incidentally that applies to basically any promise that involves "always", not just in tech.)
Wording aside, I don't think these privacy policies have teeth. If they violate it, at worst they'll lose some users and have a harder time acquiring new users who care about this stuff, which may not be a lot of users.
Therefore, I prefer not giving them easily correlated data in the first place. Sure, my VPN provider knows I'm going to google.com/bing.com/duckduckgo.com/etc., violate away, that's the necessary sacrifice of using the Internet without going full paranoid mode. Thanks to TLS they don't know the content.
These days the most vocal are GPTmaxxing, tomorrow there will be a new shiny thing.
You could buy an allocation of tokens that would be difficult to link together when used through a large VPN and fingerprint-resistant browser.
More info here https://kagifeedback.org/d/653-completely-anonymous-searches...
It’s a personal choice at the end of the day; and I aspire for a world where there is genuine choice of companies. For the time being, I’m supporting competition by using Kagi (and I get a better search engine!)
I've used kagi for 6 months and have over 7500 searches with them. It mostly works, but there are a few downsides compared to Google:
- The latency is a lot higher than google, taking over a second to display any results. - The results are often not as relevant, I have to frequently retry my search in Google. - The results for anything local (I'm not in the US) are abysmal. Searching for anything in my city instead only gives me results for the city's history.
Still, I persist in using Kagi, mainly because it's not Google and I want them to succeed. The results are frequently good enough for me to stay with them.
I'd be willing to pay up to 3$ a month for my searches, but also per-use.
If I make 0 searches, why do I need to pay?
A replacement for Google that is to survive should really convince and be super cheap, it's so easy to ignore sponsored search results (for now).
Would you use your search engine more if every search query were an implicit microtransaction? Would you use it more or less if you had to consider that your first search of the month cost $3, or if it's the last day of the month and you need to search for something but you have to wait to not incur a full month's fee.
This is one of those arguments that sounds reasonable but isn't. Nothing but a flat fee structure makes sense for something you'll be doing hundreds if not thousands of times a month. And let's be real: if you're the kind of person who could go a full month with no web searches, you don't want or need what Kagi is offering.
And one way to "deal with" that is get used to it and forget about it unless you get some surprise hit (which can be avoided with a cap). But you'll have a warm glow feel that it's "fair"
It's not like this is some novel issue average people have never had exposure to (eg, utilities)
I think it should be free or cheaper for people who genuinely cant afford it to give the equity of access that Google does.
https://files.horizon.pics/dfcedf80-2b55-422e-9960-42d730532...
The market for most is met with ads. It's why streaming services are adding ad tiers and JCPenney doing "away with constant sales and coupons, opting instead for everyday low prices" failed [1]. That's most consumers. It's almost all non-premium consumers. That's good fodder for Google and whatever LLM garbage replaces them.
Paraphrasing Scott Galloway, advertising is a tax on the stupid and the poor. I wish something like Kagi got public funding. But we have better priorities than taking ads out of search. So for the time being, you get one product for the wealthy and savvy and another, that's just good enough, for everyone else.
[1] https://excelsiorcapital.substack.com/p/jc-penneys-lost-barg...
Kagi simply does what you ask and stays out of the way. Because you pay them, they don't need to monetize every pixel on the screen. No tracking or data mining. It's just software that does what it says it does and does what you tell it.
It's tragic that "thing that does one thing well and doesn't spy on you" is a scary and alien concept these days.
This seems like a hard problem to solve, the incentive to be top ranked is just so high. What could be the solution? Can AI even help? Do we need to go back to manual curation after all? I remember in the 90s there were manually curated lists of websites, something like a website directory. At this point I'd rather get recommended a list of websites from a reddit user than relying on Google's ranking.
Part of the situation is that a company that relies on ad revenue will get gradually feel the pull of the advertisers more and more.
I'd be more worried about someone nefarious buying Kagi if it got big. Someone else would be willing to pay a whole lot of money for those eyeballs.
Indeed. And perhaps part of the issue is that there is not a single solution.
Even manual curation is ultimately based on trust. If someone's trusted list of recommendations gets popular enough, what's to stop them from "selling out," breaking that trust, to make money?
Also, the good curated stuff is typically correspondingly small and focused. But lots of people want broad results in their searches, and it's hard to imagine a person or handful of people being able to cater to all of those varying needs equally well.
Sometimes a person wants excellent narrow results (eg: academic looking for papers), other times they want broad shallow stuff, and at various other points want various other things in between.
There's a whole field of expertise, sometimes called "Library and Information Science" about organizing and making information findable, since long before computers existed. Even for them it is not a solved problem.
But the cat-and-mouse arms race that the online version has turned into makes libraries and asking a librarian for recommendations feel a lot more appealing (:
To not be top ranked ? /s
Kagi, at least for now, is making its USP the fact that it surfaces more professional, curated results. Its algorithm is susceptible to manipulation, for sure, but unlike Google, it actually has an incentive to keep SEO garbage off the first page of results.
Lo and behold it is exactly the same as Google pagerank.
2NE1 ages
I can go through this list and see you are exactly the same, except you leave up more spam.
Just one example of thousands I've looked at.
Also if you believe this is wrong you should submit search quality feedback to kagifeedback.org
We get a lot of feedback but it is mostly for technical queries that we usually address:
https://kagifeedback.org/t/search-quality
If you provide details what went wrong in kagi results for this query (and what sites should rank #1, #2... in your opinion) we can take a look. With search quality because it is such a broad space, what does not get reported, does not get looked at and addressed.
I think you need to take a good hard look at what makes for shitty content and build some parameters off that. And if you had to go off pagerank (to begin with) I would be trying stuff like adding hidden penalties to popular CMSes, boosting reddit/HN content, and following SEO trends just to thwart them. I would categorize websites by expertise so queries related to korea do not rank pages from the hindustantimes. When you search for air filters, you should probably get that housefresh team and not forbes etc. The upvote system you have is moving in the right direction but even that will need to be fortified with anti-seoer measures.
I would try to create real EEAT standards that cannot be gamed without massive investments.
It's too late to undo the damage that the Danny Sullivans of the world have done but maybe can save something here.
Reeks of astroturfing.
It felt truly bizarre to subscribe to a search engine. To actually pay for access. There’s been a bit of drama with the CEO directly emailing people when they left poor reviews.
Some people are not happy with Kagi investing in browser development instead of search results quality.
I’m not surprised there’s a lot of people having thoughts and feelings about Kagi and expressing them. The fact that there’s a significant overlap between HN and Kagi’s user bases is hardly a surprise either.
As someone who uses Orion as their daily driver, I'll admit I'm somewhat confused by why Kagi isn't staying mission focussed. That said, it may be the case that they're a premium company for a small, well-defined niche. In that case, broadening the service offering makes sense--it's what Apple did.
The typical kagi user uses hn. In the past, hn was even further up, though I guess they're slowly getting "normal" people too.
Is it? I see paragraphs of testimony from HN users who love anything whether that's a programming language, web framework, etc.
Especially if you have used kagi
But the truth is that subscribers are happy because it’s the only decent alternative out there. Google/DDG/Bing all suck. SearX may be good and free but I haven’t tried it yet.
- there are other search engines, like kagi
- we have AI, chatgpt, which some people find use for. I use it to ask general questions about programming problem solving and it often helps me. I do not know every language, and when I needed to write browser extension it saved me a lot of time
- bookmarking software. I wrote my own self-hosted software [1], with search ability. I do not have to 'research' everything I find interesting. Google will not bury inconvenient news anymore, or filter it
I think we need is adapt. When Google entered, Yahoo was not able to change. IBM was not able to change. Old tech is replaced by new solutions. We need to reconsider how to find things online.
99% of people search less than 10 times a day with loading the next page counting as a search. That's interesting and hard to accept but might be true with many living in apps and wall gardens like facebook. I wonder what the mode number is for those 99%. 1 or 2 searches a day.
There are also a fair number of queries that are just "Facebook", "<name of famous newspaper>" etc. which would also probably count towards some of the quota; it'd be great if some sort of caching could be implemented for these.
There would be fewer of these queries if people had a financial incentive to learn to use bookmarks and the address bar.
that's pretty crazy to hear. Especially since on this day alone, a weekend, I seem to have 20 unique searches. I can easily hit triple digits a day when researching for a project or on the job. Search is invaluable to me.
This is mainly because “free” search engines suck at it these days, and I’m curious if Kagi doesn’t.
I search for Errors / StackTraces that I get. For me stackoverflow / reddit / forum answers are often more helpful.
Or examples on how to implement something, the documentation can sometimes be a bit lacking on how to set things together. Give me some working code that I can fiddle with.
High level comparison between two frameworks / libs that I'm not familiar with.
Typically I’ll go directly to the online documentation from one of my bookmarks. Sometimes I’ll have a local version. I did use to just type the thing I was looking for into the google search bar in Firefox, but once the results started being for ridiculous articles (or similar), rather than the actual documentation I started using bookmarks. Which was sort of why I was curious.
> the documentation can sometimes be a bit lacking on how to set things together
If you have the time I’d love to see an example of some random person on the internet giving you a better introduction into using a language library than the documentation itself. Don’t think I’ve ever seen that.
That being said, I think we simply work on very different things. I’m not sure what searching for an error in my code would help me achieve that reading the error output wouldn’t. I suspect this is because you may be stringing together a lot of frameworks and possibly higher level external libraries, that you’re perhaps not too familiar with? Which would also explain why the documentation you have to work with isn’t always very good.
Still works for me. When I search for "Angular signal documentation" I get to the right place (They just changed domain for the brand new version, but .. yeah) That said I also use an ad blocker.
Currently google still works for me.
> If you have the time I’d love to see an example of some random person on the internet giving you a better introduction into using a language library than the documentation itself. Don’t think I’ve ever seen that.
Can't remember a good example right now. Most recent trouble that I search for was the good old classic of centering a div in CSS :) I think I used a mixture of this side [1] and an LLM ( Github co pilot)
Uh, now I found a small example. When I want to know how to sort a stream in java. When I search for "java 17 stream" go to the official documentation and search for sort I get: [2] more or less it just says "Stream<T> sorted(Comparator<? super T> comparator)"
But when I google "java stream sort" and the first stack overflow contains a great example: [3] or a bit short answer on the same page [4] Those code examples just work better for my brain :)
[1] https://www.w3schools.com/csS/css_align.asp [2] https://docs.oracle.com/en/java/javase/17/docs/api/java.base... [3] https://stackoverflow.com/a/53183266 [4] https://stackoverflow.com/a/40518343
I use LLMs, almost exclusively as fancy auto-complete because I’ve never had a computation result from them that wasn’t wrong. I think I would frankly ask one before I use Google though. I used ChatGPT for a recipe the other day. For baking very basic bread which I sort of know how to do, or at least well enough to spot a terrible recipe even though I can’t do it without one. The three results I clicked on Google were worse than the chat bot, though to be fair, two of them were just advertisements and probably written by an LLM.
and of course: discussions among communities talking about all of the above. Be it benchmarks, landmines to look out for, bug reports, highly opinated design choices, etc. Definitely couldn't find all this restrained to a Discord server or Facebook page.
And that's just all on the business end. Sometimes you just want to search up a reaction gif for a chat, or find news of the goings on (which is down on the weekend).
https://help.kagi.com/kagi/features/lenses.html
I do wish they’d increase the number of domains allowed in custom lenses and the number of lenses you can create.
I didn't know we had such limitation. Have you considered creating a feature request on kagifeedback.org? This sounds like an easy fix.
Let's say you're a plumber working at a job site where your company is digging up and re-doing lines. You come across a piece of material from the old contractor that you don't recognize, so you look up the name and code printed on the material... on a search engine.
Lots of people hardly use their webbrowser, spending almost all their time within apps.
YouTube and TikTok have become popular search engines, partly for this reason.
If you only count people for whom paying a monthly subscription for search wouldn't be completely ridiculous, I'd guess these numbers would skyrocket.
I feel respected as a customer and I feel like I'm getting great value from the service. A LLM-fueled storm might be brewing on the web, but I'm definitely riding into it with Kagi.
Funny, in terms of (non-niche) brand loyalty and evangelism, I'd say it's up there for me with Apple and Delta.
There’s been some… ego-fuelled decision making happening in Kagi’s communication.
There was a time when you simply expected to pay for anything that had value. Whether that was a newspaper, a magazine, a movie, music, or even an online service like AOL or CompuServe, you paid for it and you expected a certain level of quality in return.
In the early days of the Web, it wasn't clear at all that sites could pay their bills with advertising. Then in the mid-'90s, Ethan Zuckerman invented the pop-up ad (he's apologized since) and things progressed - or regressed, if you prefer - and we slid down the long slope to companies selling your data, hyper-targeting ads, and worse.
So many of society's ills right now can be traced to the ad-driven model. It's why clickbait is lucrative, why it's more profitable to run a populist site filled with misinformation than a trustworthy news org, it's how scammers and spammers are incentivized to flood social media sites with slop.
I'd love to see Kagi succeed, and others to follow their lead. I'd much rather spend an extra $20, $30, even $50 a month or more to subscribe to a bunch of ad-free sites that I can trust than to get it all for "free" at the cost of ads, data mining, and scammy clickbait.
Perhaps for news, or video content, or music, or something else. What else do you use today for free that often feels like a Faustian bargain?
As a random aside, If Google released a paid version of their search engine, would you switch back?
Well, up until recently. Now I'm paying hundreds of dollars for a cheap copy of VSCode and I'm really not sure why I shouldn't just use the free version.
Not if Google continue to make the bulk of their money from ads.
Magazine ads, at least in specialized pubs like computer magazines, were actually useful. As a kid, I subscribed to Dragon Magazine because I played D&D, and the ads were half the fun for me.
You'd think that with the hyper-targeting of online ads, that we'd be happy with them, too. But the actual products behind them are usually low quality spam. And we've lost a lot of discoverability from all this targeting. Sometimes it's good to see things that aren't aimed directly at your immediate demographic.
The worst bits, which aren’t even that bad:
The maps are a work in progress.
No shopping search. If I want something like shopping search I can use Google.
Not sure if it’s because I’m in Australia but sometimes it takes a while for the page to load. Subsequent searches are faster.
What is it for?
For electronics I know 3rd party services that compare prices for different shops. But in cases like "Where can I by this specific dental floss" I sometimes use google shopping search.
This is like being miffed that Yellowpages doesn't pay businesses to list them in the phone book.
https://blog.kagi.com/small-web
How would you design a system where creators are paid in a search engine? We have always had an idea of profit share for showing up in results but we would need to reach about 10M members for this to be more than few cents a month. (Kagi CEO here)
I have some months where I search for stuff I already know. Like I goto url bar and just type imdb which then initiates a search. I know the dang url but am lazy lol.
I wish the team tons of luck, I monitor the pricing page monthly usually, just hoping to see a small uptick in starter limits as unlimited is just not my use case. $8.00 for unlimited and again, I would sign up just to support them.
Also it's not clear if the numbers used in the article omits publisher ads from Adsense.
It would be great if G's advertisers were opened up to competitors, I think a CPA model would work out well since combatting click fraud involves a lot of profiling.
But as with all of these things, I'm paying to no longer be the product. That's the value for me, and so "it not working exactly as I expect 100% of the time" is actually a small price worth paying.
Generally speaking, I find the quality of results usually on a par with Google.
How is this better than using Google with an ad blocker?
—-
That’s my current solutions, why should I switch to paying for Kagi? Ty!
Rubber duck to a battleship.
The number of times I've found something in seconds that a co-worker was digging around in pursuit of for minutes has by this point escaped me. Were you around for Alta Vista vs. Google? This feels the same. The only difference is it's paywalled, which for the consumer, is generally good--it means the benefits won't be generalised and the product will remain an elite minority offering that doesn't gain traction with SEO bots.
I like that I get to choose when to use GPT.
If you’re on iOS, the Orion browser is great, too - tldr chrome extensions on iOS.
The quick summary is that they aren’t really privacy focused at all, don’t comply with GDPR, and generally seem ok misleading their customers. The lack-of-privacy doesn’t bother me itself, but misleading customers really troubles me. Why would I trust them with X when I know they’ve lied with Y?
The article you point to is a perception of one person, unfortunately sharing screenshots out of context to create a negative narrative and exploiting a fact that we run a 100% transparent business where you can ask the CEO literally anything directly.
Our policies have always and will belong in kagi.com/privacy and not on random websites. And it is hard enough to do what Kagi does even without misinformation shared in bad faith, so I'd ask you to please not do that going forward.
Here is why Kagi exists:
I notice that your GDPR policy is not discussed on the policy page you provided. Here’s an archive taken just now: https://archive.ph/9bnhY
> If you are not paying for it, you’re not the customer; you’re the product being sold. -- Andrew Lewis
Also, plently of people at Google, Meta, etc feel they are doing good by providing free ads based services.
I suspect people with low income are way better off with free search, free maps, free docs than if each was $3 a month. It doesn't sound like much but their were definitely times in my life where $3 a month per service felt like too much. And yet, those services arguably provide extraordinary value for their users.
For national searches, I find Kagi good at long as you specify the country. That’s one of my favorite features actually, you can leave it international by default, and add “!es” to search in Spain.
The good part is that after you've selected your country as the region, you still get good results in English or other languages.
Half the websites it shows are those I wouldn’t have found on google and are relatively high signal for what I’m looking for (including very niche blogs from experts from the field).
Kagi integrates GPT 4 and other LLMs, as well as its own Quick Answer product. The Quick Answer product is 9 times out of 10 superior to any of the pre-trained LLMs. Mostly because it's accessing live information.
Put another way, I'd love to go head to head with a competitor who relies on ChatGPT for their queries.