Google.com is REALLY bad right now in my opinion.

The AI answer dominates the page, and is usually wrong, so it's just a waste of the most valuable space on the page.

The ads have proliferated, and rarely have what I need.

Video search results also take up a lot of the page, and are almost never what I want if I haven't clicked over to the Video tab. Again, just a big waste of space.

Speaking of the Video tab its accuracy/relevancy has gotten worse too.

The organic results have deteriorated, and are still competitive, but are no longer clearly and consistently better than Duckduckgo.

So it's duckduckgo as the default for me on all devices now. If I don't find what I want, which is maybe 30% of the time, I switch over to G just by adding !g to the query. G might have what I need in about half of those cases.

You can see why a lot of people just ask an LLM to do the web searching for them.

Across the board, this is all quite bad compared to web search from 10+ years ago.

If there's a single thing which has really pushed me away from Google, it's simply google.com devoting so much of the page to things that aren't organic web search.

This is similar to how, not too long ago, LLM's had extreme difficulty counting the number of letters in some words. LLM's don't "think" or "reason" in the normal definition of those terms. They can do some pretty amazing things, but still screw up basic things like telling you something that is obviously wrong and contradicts the top search results.

LLM's, in their present stage of development, are sort of like a crack-addled idiot savant. Sometimes they are obviously insane, and sometimes they seem quite cogent, but you must never trust them implicitly. This may be why they are so difficult to constrain. You could give them something equivalent to the laws of robotics, but following laws requires thought processes they simply don't have.

I'm actually sort of amazed Google doesn't make people accept some kind of butt-covering EULA and post disclaimers about the inaccuracy of results before even showing you their AI's output. Are they not being sued over this kind of thing?

The AI is just signal to the market that they are AI first. Their AI on search is pure garbage. It's obviously a very super light weight model to be able to support the volume of daily searches. If they used their frontier to power search folks will not pay for their cloud offering.

Google is no longer a search engine anymore. They don’t care about being one. Why you might ask?

First, a search engine indexes the web and makes it available to users. It’s been ages since Google did any of that. They no longer index sites or take ages to do so. Case in point, our cybersecurity startup (Webvetted.com) was launched in November 2025. Till date, only one page is indexed on the entire website. And I’ve talked to lots of other developers and it’s a common issue.

Secondly, a search engine organizes indexed information and makes it useful for people. Google is a basic LLM nowadays. They figured out that why organize and make the information useful when they could just answer the question with Gemini anyways? So they no longer bother to do the work of a search engine and are now just a lower-ranking open-source Chinese LLM

I think you overestimate how much people want an actual truthful answer. I think people are often looking things up to silence a nagging voice in the back of their heads, or settle an argument to be able to move on. In those cases, no decision is riding on the answer, and a quick answer, even when factually incorrect, could be argued to serve the purpose just as well as (or better than) a slow but more correct answer.

I realised this for myself when I read Sivers on being without internet[1]. He discovered a process to silence the nagging voice without searching:

> When I’m yearning to search, I ask myself why.

> - What answer am I hoping to hear?

> - What answer would be a surprise?

> - What would I do in each case?

[1]: https://sive.rs/off23

I similarly noticed Gemini absolutely refuses to look at a url when I give it one and will instead just hallucinate based on what it thinks the url is. Here I am assuming Google will have the best web capabilities in its AI.

It is using the search engine. The AI summary is run over the top 100 results for your query (or something)

The problem is they're just running a very dumb, cheap model on the results because running a smart model on every search result page would cost them infinity money.

The other day, it was the Aussie AFL final. We were in the car and asked it for a score update.

"The team are tied at the end of the third round with the scores 43 to 51".

So it was tied at the end of round 2 for 43, not round 3. Not sure were it got the 51 from and why it figured it was a tie. LLM's pretty cool until they aren't. As they say, the hallucinate 100% of the time but most of the time it is useful.

Make the switch:

Search engine > DuckDuckGo / Qwant / Ecosia

Email > Tuta Mail

Photos > Ente

Cloud storage > Nextcloud / Internxt

Office > CryptPad / LibreOffice

Maps > OpenStreetMap, OsmAnd

Operating systems > LineageOS (mobile), Linux (Ubuntu, Fedora Debian)

Calendar > Nextcloud Calendar / Tuta Calendar

Google will stop being a traditional search engine because they believe they’ve found a more profitable version of it.

One where their users don’t go off to other sites and where they can keep shoving ads in their face.

How did you google this, and did you use the dumb search, or specifically 'AI mode' lens?

Because I did this and got a vastly different result from you:

Yes, the Halifax Wanderers can still mathematically qualify for the 2026 Canadian Premier League (CPL) playoffs.The top four teams advance to the postseason. Following their 1-0 loss to Atlético Ottawa on September 26, 2026, the Wanderers sit in fifth place, just below the playoff line.

With a indexed table of the games and the playoff table, with a breakdown of what the points they need to achieve to do so.

The search window may not always crawl sources. AI mode specifically does some research before giving you a response. Not sure what you're on about.

It's meant to be A SEARCH ENGINE

I suspect for a long time now it's ultimately meant to be a revenue-maxing ad platform.

Google is horrible now and search doesnt show anything relevant anymore. I miss old Google, with 30 pages of links. Now if you search for a specific quote and it wont find it. Everything is walled now and content you get is the one they want to push.

Commercially, this may be what they want to happen. Instead of just getting an answer, they got you to make three more prompts / queries. What do you know? "Engagement" just went up 3X!

Kagi works like you would expect. It searches first - and then if you’ve ended your search with a ? or configured it to always do this - it passes the search results into the assistant and gives you a summary. You can change the default model used for this if you think the cheapest, fastest model Google has is not good enough for the rare occasions you want any model’s opinions about your search results.

> My question is: what's the point of the AI in the search engine if it itself isn't going to use the search engine first before answering?

AI is sold well because the user either corrects it or happily accepts any answer (usually, depending if one is an expert in the question's field).

Search is not the google's area, and it is not the search they sell! They sell ads.

I feel like this might be self-preservation of most of the frontier model companies in that at first they wanted to show off the models with all the cool actuators that it had at its disposal, but then when profits weren't keeping up, they nerfed them to need to be specifically asked to escalate levels of effort, which the Google search AI isn't explicitly told to do in its pre-fetch.

How about Alexa - given a "6 minute green beans timer" - later reporting 11m remaining on said timer? Confronted with the error, "You're right! That 6 minute green beans timer was too ambitious."??!!

I can't even.

Searching first is how Kagi assistant works and it's great.

I remember when Google was accused for forcing users to go through several pages for answers, because this translated into more clicks, and higher engagement. I am curious whether AI overview will be as bad, as people can tolerate for you to spend more time with it, more queries, which translates into more time spent with Google.

They can show numbers, how much people use their engine, and the numbers goes up.

What do you mean? I can guarantee that there would at least be 3-4 people who got their promo packets approved for this feature. Google’s features exist for its employees to be promoted.

google.com is like the guy in the old joke:

- What's your best skill?

- I can count really fast!

- What's 432*567

- 912

- You sure? Sounds off!

- Hard to tell, but it was fast, wasn't it?!

I think they're training their models using us correcting them repeatedly. That's my tin foil hat theory and I'm sticking to it!

After all, why would Google do anything for free when it comes to AI?

If you wanted to tease customers with one AI shitty free search box in order for them to then decide to upgrade and pay for Gemini, well, this isn't the way. And so either Google are idiots, or we're helping them for free. I'm going with Occam's Razor on this one - we're the product.

I had an incredibly dumb response today too. Appeared to fully understand the question, just 100% factually incorrect answer. https://mstdn.social/@sgraham/117346578326361464

Until Google wants to pay for reasoning for everyone their AI results will continue to be terrible. In my experience, reasoning fixes a great many of the sorts of hallucinations and falsehoods I've received from googling.

This is the reason why most people underestimate AI. These free to use models can do a very poor job in easy things. Some people assume that’s the state of the art with AI.

On current website data and very recent information, Gemini is explicitly forbidden from acting as a search engine; If you ask it for references specifically they tend to be hallucinated, over and over again, until it pleads "Sorry, I can't access the Internet".

I'm being very mindful of Gell-Mann Amnesia and chatbots. They speak authoritatively and are right often enough, but I've had enough cases of them saying very incorrect things in areas I know well that I have to remind myself that those cases aren't unique.

I find that it is exceptionally bad with sports. I don't really want to google anything about the Texas A&M Aggies at the moment, but when I did, it would regularly hallucinate scores, hallucinate opponents, get the locations of games wrong, confuse A&M's quarterback with someone else, etc. Maybe its better in other domains; I don't know, I mostly don't even use Google these days, pretty much entirely because I want to avoid the AI Overview, which is just wrong way too often to be the first result.

Exactly. Makes me wonder how much of the compute tax on energy would be saved from simply reverting google search to default (the old way)

I guess the AB tests say people want fast and confidently wrong more than they want slower and correct.

while i share your frustration, Google for their part is trying to adapting a new technology into an existing service. while the whole endeavor might have been misguided, not sure we want to blame them for trying...

free tier AI is still horrible, the responsible thing would be for search engines to stop serving it at all, but somehow they must think they will make more money misleading people so they mislead people.

such is the nature of public companies.

"Gemini, is today the 3rd Thursday of the month, I need to know if I can park here or if the street sweepers coming" No you can park there. "Is Today Thursday?" Yes "Is Today Wednesday?" Yes

This seems like an implementation bug.

They are probably caching the answers. AI-answers are most likely never going to be profitable even with a smaller model.

You are posing a question that requires deductive reasoning. LLMs can only approximate that if they have sufficient corpus to find their way to a believable response. Their willingness to spout bullshit in such scenarios where their sources are sparse is a problem but you can't expect them to function like a fully developed mind.

The gogol agenda of blocking and excluding all web engines not part of their whatwg cartel is moving forward, at a slow pace, steathly.

gogol search was blocked to all noscript/basic HTML browsers last year. I recall people that a few years before, they removed their gmail basic HTML interface not long after that their account creation (then login) would require one the abominations of the whatwg cartel.

I have been working around the gogol search engine since then and my email is now self hosted without DNS (IPv6 literals, see RFC). Yeah, most DNS registrars are now gated by the whatwg cartel web engines in some way, this is the net effect of big corpos toxic behavior.

Pure evil.

The first time, sure, it saves compute. The second time you ask, I feel like it should be the time to go into thinking/verification mode. But who are you? Are you paying? Does the answer being correct generate ad dollars? No? Then your usage mode isn't even being optimized for in their A/B test, probably.

In fact, if hallucinating the wrong answer hooks you into doing even more searches or into buying something useless, it would be preferred!

The problem is one of information provenance. A search engine shows you stuff that other people have written. It might be right or it might be wrong. But you get to judge that yourself, because you have an idea of where the information came from.

With AI summaries they take that away from you and claim the provenance for themselves. Google is now giving you the answer and they hide where that answer came from. Therefore, you cannot properly evaluate the provenance of the information presented. It still might be right or wrong, but you can no longer properly evaluate it based on source.

It feels a bit like how Amazon and META purposefully do not create search that can actually help you filter for exactly what you want.

There are incentives to create products that work just barely, just enough for you to use them and don't leave for elsewhere but then they want to maximize everything else like the probability you view side products or ads or just click around and type and generate more data.

Unless someone else offers something better they will offer something that is basically as bad as they can get away with so long as there are benefits to them for doing so..

Well, isn't this an example of why Google are going in the direction the OP is complaining about? Why are you searching with a question? Are you hoping to find a forum post with this exact question or something?

I don't understand the game you are interested in, but you should be searching something like "Xball qualifying results" or "Xball playoff rules" or something. Instead, by asking a question you are expecting Google to be AI.

They know what people are typing in so it seems they are quite justified into pivoting from a search engine to an AI.

I agree. When I read the post I immediately thought what is Gemini chat doing there when it should be a search engine. It could probably give sensible answers if it defaulted to search first. It looks more like they are trying to showcase Gemini, not augment google search. They might even be thinking "search" is a dying product.

When I’m using a public urinal in 2026, it’s flushing three times, except after you’ve finished. Instead the light switches off during the act, which you’re carrying out blindfolded then. These things were more or less perfected technologies a few decades ago.

The reason I cannot spare HN has to do with the Marxist concept of added value. Enshittification becomes necessary once growth isn’t possible anymore on a corporations home turf, where it was actually innovative.

And it's going to get worse over time (kinda like how Google search results got worse over time) as ads get introduced and people try to game the ai response. General purpose AI chatbots are a waste.

They’re trying to keep it from “thinking too hard”/using “too many” resources.

That simple.

Broad rules across specific subjects like this are tough to get right.

Just lots of fine-tuning and exceptions.

I had their chatbot avoid pulling a URL from archive.org with a couple pretty impressive steps of mental gymnastics basically telling me how I could do it myself, but refusing until I pushed.

> My question is: what's the point of the AI in the search engine if it itself isn't going to use the search engine first before answering?

Because the point of AI slop is to waste your time. You just lost about 30 seconds of your life trying to get a correct answer. AI was lying to you, so you had to spend time to counter the AI slop lies here.

I solved it by banning all AI slopness; in the browser some extensions do that. The world becomes better without AI slopness.