Mullvad: Shutting down our search proxy Leta
mullvad.net
mullvad.net
https://developers.google.com/terms
> you will not [...] keep cached copies longer than permitted by the cache header
I think it's easier to just call it "lying." Nearly to the point of "breaking the law."
My guess is search's days are numbered and companies are "pivoting" away to other projects
a shutdown is preferable to silent bitrot
Right this moment it seems to work. 2 Days ago Id search for something basic like "CSS colors" and not get back a single usable result
but yeah its a combination of SEO trash but it also seemingly not having stuff indexed
I'd search "Ask Clojure" or "Clojure Agents" and nothing from Clojure.org would show up
or like "MDN SVG circle" and the MDN would just not show up.
of course today im trying it and its all working haha
re blocking: after every search in the upper-right corner of each link I see 3 dots which opens a menu and offers "block this site from all results".
In the last year or so, I look for the summaries from “Search Assist” and the dive into a chat with the (limited?) LLM models that it provides. It’s my go to for LLM usage. It’s rarely and for more complex needs that I go to ChatGPT.
Wonder what's different, it seems people's experience differ quite a lot.
Kagi is proof of this. Kagi results are almost all the Google search API. It shows that Google is completely capable of building a better search engine if they wanted to
I think YMMV though and this is probably highly location dependant.
I still use it because those are not very relevant for me, but it’s a shame.
For $10/month it’s great to have someone whose incentives are aligned with my own managing my relationships with AI companies I’d otherwise have to monitor constantly for privacy abuses.
I haven’t noticed a recent degradation in DDG’s search results, but I’m also turning to duck.ai more frequently and on the whole my search/investigation experience is better.
The one significant downside is that duck.ai limits the length of your chats, but considering the price that’s not surprising.
The direction I’d like to see the industry go is better integration of search results into AI chat, blurring the distinction between the two. That would make both products more compelling: search results are made more friendly with AI summaries, and original sources help to counter AI hallucinations and obsequious blather.
[0]: I won't bother linking any articles since there are too many articles on the subject and whatever I link is probably not the site you want (or is maybe paywalled).
My hope is that AI helps to fine tune inquiries and helps users discover websites that would otherwise not have been uncovered by traditional index-based search.
Unfortunately it’s in the interests of search and AI companies to keep you inside their portals, so they may be less than willing to link to the outside even when it would improve the experience.
[0] https://www.jaronlanier.com
edit: grammar
I almost always get the claim(1) and footnote with URL or book, or DOI.
The default internet device these days is the phone; so many people don’t even use desktop any more. Space limitations on small screens mean that this is unlikely to be shown by default. Moreover, phone interfaces discourage most users from opening multiple new tabs forking off any webpage. You might show desktop users this and get some uptake, but that’s not enough to save the open web.
This is true, but aren't "AI" summaries directly opposed to this interest? The user will usually get the answer they need much more quickly than if they had to scroll down the page, hunt for the right result, and get exposed to ads. So "AI" summaries are actually the better user experience.
In time I'm sure that we'll see ads embedded in these as well, but in the current stage of the "AI" hype cycle, users actually benefit from this feature.
I think that's hyperbole.
Yes, users can rely on "AI" summaries if they want a quick answer, but they've been able to do that for years via page snippets underneath each result, which usually highlight the relevant part of the page. The same argument was made when search engines began showing page snippets, yet we found a balance, and websites are still alive.
On the contrary, there's an argument to be made that search engines providing answers is the better user experience. I don't want to be forced to visit a website, which will likely have filler, popups, and be SEO'd to hell, when I can get the information I want in a fraction of the time and effort, within a consistent interface. If I do need additional information, then I can go to the source.
I do agree with the idea you mention below of search engines providing source links, but even without it, "AI" summaries can hardly be blamed for hurting website traffic. Websites are doing that on their own with user hostile design, SEO spam, scams, etc.
There is a long list of issues we can criticize search engines for, and the use of "AI" even more so, but machine-generated summaries on SERPs is not one of them IMO.
My issue with AI summaries is that they are not even remotely accurate, trustworthy or deterministic. Someone else posted this wonderful evidence[1] in the comments. LLMs are sycophantic and agree with you all the time, even if it means making shit up. Maybe things will improve, but for the last 2 years, I have not seen much progress regarding hallucinations or deterministic i.e. reliable/trustworthy responses. They are still stochastic token guessers with some magic tricks sprinkled on top to make results slightly better than last month's LLMs.
And what happens when people stop creating new websites because they aren't getting any visitors (and by extension ad-revenue)? New info will stop being disseminated. Where will AI summarize data, if there is no new data to summarize? I guess they can just keep rehashing the new AI-generated websites, and it will be one big pile of endlessly recycled AI shit :)
p.s. I don't disagree with you regarding SEO spam, hostile design, cookie popups, etc. There is even a hilariously sad website[2] which points out how annoying websites have become. But using non-deterministic sycophantic AI to "summarize" websites is not the answer, at least not in the current form.
[0] https://www.theregister.com/2025/07/22/google_ai_overviews_s...
[1] https://imgur.com/a/why-llm-based-search-is-scam-lAd3UHn
[2] https://how-i-experience-web-today.com/
edit: grammar
> My issue with AI summaries is that they are not even remotely accurate, trustworthy or deterministic.
I am firmly on the "AI" skeptic side of this discussion. And yet if there's anything this technology is actually useful for is for summarizing content and extracting key points from it. Search engines contain massive amounts of data. Training a statistical model on it that can provide instant results to arbitrary queries is a far more efficient method of making the data useful for users than showing them a sorted list of results which may or may not be useful.
Yes, it might not be 100% accurate, but based on my own experience, it is reliable for the vast majority of use cases. Certainly beats hunting for what I need in an arbitrarily ordered list and visiting hostile web sites.
> LLMs are sycophantic and agree with you all the time, even if it means making shit up.
Those are issues that plague conversational UIs, and long context windows. "AI" summaries answer a single query and the context is volatile.
> And what happens when people stop creating new websites because they aren't getting any visitors (and by extension ad-revenue)? New info will stop being disseminated.
That's baseless fearmongering and speculation. Websites might be impacted by this feature, but they will cope, and we'll find ways to avoid the doomsday scenario you're envisioning.
Some search engines like Kagi already provide references under their "AI" summaries. If Google is pressured to do so, they will likely do the same as well.
So the web will survive this specific feature. Website authors should be more preoccupied with providing better content than with search engines stealing their traffic. I do think that "AI" is a net negative for the world in general, but that's a separate discussion.
btw your linked article/study doesn't support your argument - SERPs are definitely stealing clicks (just not nearly as many as AI):
> In other words, it looks like the featured snippet is stealing clicks from the #1 ranking result.
I should maybe clarify: I have been using LLMs since the day they arrived on the scene and I have a love/hate relationship with them. I do use summaries sometimes, but I generally still prefer to just at least skim TFA unless it's something where I don't care about perfect accuracy. BTW did you click on that imgur link? It's pretty damning - the AI summary you get depends entirely on how you phrase your query!
> Yes, it might not be 100% accurate, but based on my own experience, it is reliable for the vast majority of use cases. Certainly beats hunting for what I need in an arbitrarily ordered list and visiting hostile web sites.
What does "vast majority" mean? 9 out of 10? Did/do you double-check the accuracy regularly? Or did you stop verifying after reaching the consensus that X/Y were accurate enough? I can imagine as a tech-savvy individual, that you still verify from time to time and remain skeptical but think of 99% of the users who don't care/won't bother - who just assume AI summaries are fact. That's where the crux of my issue lies: they are selling AI output as fact, when in fact, it's query-dependent, which is just insane. This will (or surely has) cost plenty of people dearly. Sure, reading a summary of the daily news is probably not gonna hurt anyone, but I can imagine people have/will get into trouble believing a summary for some queries e.g. renter rights - which I did recently (combination summaries + paid LLMs), and almost believed it until I double-checked with a friend who works in this area who then pointed out a few minor but critical mistakes, which then saved my ass from signing some bad paperwork. I'm pretty sure AI summaries are still just inaccurate, non-deterministic LLMs with some special sauce to make them slightly less sketchy.
> Those are issues that plague conversational UIs, and long context windows. "AI" summaries answer a single query and the context is volatile.
Just open that imgur link. Or try it for yourself. Or maybe you are just good at prompting/querying and get better results.
> So the web will survive this specific feature. Website authors should be more preoccupied with providing better content than with search engines stealing their traffic.
I agree the web will survive in some form or other, but as my Register link shows (with MANY linked studies), it already IS killing web traffic to a great degree because 99% of users believe the summaries. I really hope you are right, and the web is able to weather this onslaught.
I don't verify the accuracy regularly, no. And I do concede that I may be misled by the results.
But then again, this was also possible before "AI". You can find arguments on the web supporting literally any viewpoint you can imagine. The responsiblity of discerning fact from fiction remains with the user, as it always has.
> Just open that imgur link. Or try it for yourself. Or maybe you are just good at prompting/querying and get better results.
I'm not any better at it than any proficient search engine user.
The issue I see with that Imgur link is that those are not search queries. They are presented as claims, and the "AI" will pull from sources that back up those claims. You would see the same claims made by web sites listed in the results. In fact, I see that there's a link next to each paragraph which will likely lead you to the source website. (The source website might also be "AI" slop, but that's a separate matter...) So Google is already doing what you mentioned as a good idea above.
All the "AI" is doing there is summarizing content you would find without it as well. That's not proof of hallucinations, sycophancy, or anything else you mentioned. What it does is simplify the user experience, like I said. These tools still suffer from these and other issues, but this particular use case is not proof of it.
So instead of phrasing a query as a claim ("NFL viewership is up"), I would phrase it using keywords ("NFL viewership statistics 2025"). Then I would see the summarized statistics presented by "AI", drill down and go to the source, and make up my mind on which source to trust. What I wouldn't do is blindly trust results from my biased claim, whether they're presented by "AI" or any website.
> it already IS killing web traffic to a great degree because 99% of users believe the summaries. I really hope you are right, and the web is able to weather this onslaught.
I don't disagree that this feature can impact website traffic. But I'm saying that "killing" is hyperbole. The web is already a cesspool of disinformation, spam, and scams. "AI" will make this even worse by enabling website authors to generate even more of it. But I'm not concerned at all about a feature that right now makes extracting data from the web a little bit more usable and safer. I'm sure that this feature will eventually also be enshittified by ads, but right now, I'd say users gain more from it than what they lose.
E.g. if my grandma can get the information she needs from Google instead of visiting a site that will infect her computer with spyware and expose her to scams, then that's a good thing, even if that information is generated by a tool that can be wrong. I can explain this to her, but can't easily protect her from disinformation, nor from any other active threat on the modern web.
Who cares if it's deterministic? Google changes their algorithms all the time, you don't know what its devs will come up with next, when they release it, when they deploy it, when the previous cache gets cleared. It doesn't matter.
That is a big if.
A summary cannot be better than what is summarizes in any was but brevity. It can be much worse.
Frequently I cannot even find source documents which match my exact circumstances; I’m uncertain whether they actually exist.
To be honest, DDG has always been far behind Google. It’s fine when I know my search result is going to be in the top 10 of any engine I use, but the moment I need to search for anything non obvious I don’t even bother with DDG any more.
DDG does seem marginally worse today than it was maybe 5 years ago. It falls off rapidly past the first few results. Now it even seems like it just starts mixing generic results from some popular adjacent keyword into the results and hopes we don’t notice as users that it stopped trying to search by page 2.
And so, even if Google was the same thing it was back in 2010, there's no longer anything for "search" to find. And I hope you all downvote me to -50 and scream at me for being a retard with some snarky-assed abuse detailing how and why I am wrong. Because I don't want to be correct about this.
There was an era where there were a lot of completely free sites, because they were mostly academic or passion projects, both of which are subsidized by other means.
Then there were ads. Banner adds, Google's less obtrusive text ads, etc. There were a number of sites completely supported by ads. Including a lot of blogs.
And forums. Google+ managed to kill a lot of niche communities by offering them a much easier way to create a community and then killing it off.
Now forums have been replaced by Discord and Reddit. Deep project sites still exist but are rarer. Social media has consolidated. Most people don't have personal home pages. There's a bunch of stuff that's paywalled behind Patreon.
And all of that has been happening before anyone threw AI into the mix.
Pretty much. Most (all?) search engines have basically stopped indexing the web. If you create content that doesn't make it through social media and has significant links, Google won't just index your website.
No, it's not under-ranking your site. It's plainly not indexing it. So if you have weird, specific content out there; it simply won't show up for a particular search.
Search is pretty much over and no one is interested in getting that fixed.
[1] Ask HN: “Tank man” image search blocked on Bing and DuckDuckGo 560 points by MaxHoppersGhost on June 4, 2021 | 126 comments | https://news.ycombinator.com/item?id=27394925
There are more or less valid arguments for not excluding Yandex[1], but as a European, I want to avoid any of my money going to Russia if possible. And there is no setting to exclude Yandex from your Kagi search results.
If you stopped using duckduckgo because of the tankman fiasco, maybe you should reconsider if Kagi is right for you.
[1] https://kagifeedback.org/d/5445-reconsider-yandex-integratio...
EDIT: Apparently not https://law.stackexchange.com/questions/103256/can-you-use-y...
This is not like trying to use Bing, and then half the time you have to do the same search on Google because of how poor Bing's results are. It feels like Google felt fifteen years ago: useful results without all the "sponsored links" garbage around it.
Originally I switched due to their environmental focus and the way they run the company, but the quality of the results keeps me there. They have their own ! query, like DuckDuckGo. So !maps for Google Maps and !w for Wikipedia.
> We, and our search partners, collect your IP address in order to protect our service against spammers trying to conduct fraud or to up-rank specific search results.
Src.: https://www.ecosia.org/privacy
This shouldn't necessarily stop anyone; I think it should just be mentioned when it is suggested as an alternative to DuckDuckGo. You probably wouldn't switch from a search engine that proxies all favicons to avoid tracking to one that sells your identity to Google and Microsoft for tree-money.
Disclaimer: I work at Brave
Im expecting a future where we dont have “pages” on the internet anymore, but its just the backbone for generative AI content and if you want to promote your brand you need to pay the AI providers to put your content in responses.
Eventually, the entire notion of “searching the web” will be seem as archaic as the rotary phone.
It doesnt require many resource and would be easy enough to run it on docker compose alongside a valkey/redis instance. I have mine on k8s but i dont think there is a helm chart easily found.
Privacy is an uphill battle, we should use our efforts where they make the most impact.
It is the exact opposite for me. Everyone hates the LLM based search products in my circles. Just look at this shitshow.
switching to LLM based search products in droves.
The shift towards LLM-based search products is significant as they offer more conversational and personalized responses compared to traditional search engines. This change is driven by users seeking quicker, more relevant answers and a better overall experience
Yes it's meant sarcastically. Personally I agree with you that you want to look at pros and cons, both sides.
But there are many, oh very very many people out there who would literally do exactly what I wrote. They do it all the time. That's why we have these echo chambers everywhere. And AI as we can see here is not making it better.
And it's not just the "general population". I see it on a technical level too at work. Developers just trusting the AI output. It sounded confident when it said it found the root cause, fixed the bug and added tests. So it's good to commit, right?
I don't follow NFL viewership numbers but searching and reading the results seems to indicate some support for that trend.
What's wrong with it?
The LLM responded in the affirmative to all queries, even when they seem to contradict each other.
> NFL viewership went down? Yes!
> NFL viewership went up? Yes!
> Home prices are going down? Yes!
> Home prices are going up? Yes!
Google is confirming people's biases on an industrial scale. Surely this is not going to do any damage...
1) don’t believe everything you read on the internet
2) it looks like real search results
3) it’s a bunch of crappy smartphone shots of screens
4) but why would someone work so hard to make these fake images
5) but AI image generation
6) but they don’t look AI-generated
7) but maybe it’s gotten better
I can go on and on and on
My ultimate feeling is “this looks legit.” But man. The internet just isn’t fun anymore. It’s so much work.
[0] https://techcrunch.com/2025/08/25/ai-sycophancy-isnt-just-a-...
Searx is a similar idea. More powerful but definitely uglier
Mullvad sell a VPN and privacy-focused browser so how are they unable to proxy the searches themselves? They already have the needed tools developed.
If they crack down on it then the suggestion to use a VPN and privacy browser won't work either.
Still hope Russia looks like Berlin in '45 soon but until then, you can't deny they got the better search engine.