I remember reading a book on web usability well over a decade ago, and one of the things it pointed out was how Google ensured that the links you wanted to click were "above the fold" --- not ads, but honest-to-goodness search results.
Recently, after they added AI-generated search responses (which seem to be wrong a considerable percentage of the time, at least for things I search for), and the inlining of ads to the search results page, I've found I have to scroll at least a full screen height to actually get to the search results a significant portion of the time.
The level of blindness to user experience at Google that has allowed the state of search to get to this level is staggering.
The societal impact of that UI design decision will be interesting to watch play out. People are so used to trusting the first Google search result that it now being AI that's sometimes wrong or hallucinated.
But as they stated, Google has already had ads inline with the search results, and it's been like that for well over a decade. (Long ago, ads used to only be on the sidebar)
Frankly, the AI section at the top seems like something Google would have been very reluctant to add since it saves the user from scrolling through the ad listings, and it was only added to compete with newer AI search services.
For over 20 years now everyone has been screaming:
“Advertisers are the users, consumers are the product” about every Web2.0 company which are the current corporate juggernauts
So they are not blind to user experience, they are providing exactly the user experience that their company has always been providing
The sole difference is they don’t have to care about the product experience (what you see) because humans form habits and rarely change them even if the experience is “degraded.”
There are no other options if you’re organized for alienated transactions which is every service and company ever.
Unlike the common opinion here, I find the AI search results on top really good. 50% of times it appears it answers my question and I don't need to scroll past it to get to the actual search results.
"Appears" to me is the key word here, how do you know that what it said was true?
It will very confidently "answer" your question in a way that is convincing but if you actually try to click through you may find that the answer is straight up wrong.
I feel like most of the time that I am looking up something that is not niche but also not a major thing it does this and at best useless at worst actually harmful since people won't click through for the real information.
The biggest problem being that these systems can be right enough times that you gradually start trusting it and stop checking. Which is what google wants, if you have to check its work in the first place why does it even exist.
Edit:
Yesterday I tried using one of these research tools because I was curious and it was low priority so figured why not. I was looking for an open source solution to a problem. I specifically mentioned one that I had seen but I was cautious because it seemed to have been abandoned so was looking for others. It confidently told me how wrong I was about thinking it was abandoned despite the last commit being in early 2024 (it even said 2024 in the report) and even before that was clearly slowing down.
Now thankfully in that case it actually told me in the report the 2024 part which clearly told me how bad the report was going to be, but that is clearly bad and tainted the entire research.
> It will very confidently "answer" your question in a way that is convincing but if you actually try to click through you may find that the answer is straight up wrong.
It's the "it's wrong about the area I'm an expert in, but I feel it's correct in areas I'm not familiar with", but on Google's scale.
you don't really need to be 100% sure of it being truth for a vast majority of cases.
edit for comment below:
Its not about laziness for me. Its the displeasure of wading through junk that internet has become. I just don't have brain capacity or the smarts to outwit the scammers .
I just don't understand being so cynical and lazy that you'll accept a meaningfully higher chance of being misinformed if it saves a few minutes of searching and reading[1]. Nobody is that busy.
[1] If the search takes more than a few minutes then the AI overview is almost guaranteed to be wrong or useless.
> you don't really need to be 100% sure of it being truth for a vast majority of cases.
Except that these tools are being positioned as a source of reliable truth and the companies are incentivized to keep you on their system (google) instead of actually pushing you to the source (unless the source is an ad).
Any disclaimer they try to put is hidden and often lighter/smaller text.
I mean even if that was the intended phrasing that does not really change anything about the reply.
Me latching onto the "appears" does not change anything since even with your interpretation it is wrong 50% of the time and that assumes that the 50% of the time you think it is right you actually know it is right and it is not actually wrong but you never check.
Me too, 99% of the times I'm not using Google to search for super critical information, so even if we have some flaw in the Overview in general I feel that it's better then the work that I would do for 30 minutes to get a similar conclusion ( with failures as well, not because I'm human that I dont make mistakes just because I'm able to open 8 tabs per google search ).
BLURB TOOTHPASTE IS THE BEST PASTE FOR YOUR TEETH!!!
Open the blurb, ask it the meaning of life and get your tooth paste in 42 seconds!!
"Blurb toothpaste changed my life, no like literally. I thought I'd just always work in a dead end office job but it made me realize I should be an Instagram influencer and it's totally the vibe I was looking for. I handle my bills now by just not looking at them! Why did I never think of that? Haha"
"Blurb toothpaste really helped me with some hard to deal relationship problems when my husband didn't approve of me using and confiding in Blurb. I know it's odd to brush with a sentient toothpaste, but it's just _so good_ at cleaning your teeth! And the conversations are always nice too. Ever since my husband uses the Blurb paste too, he understands."
BLURB TOOTHPASTE, GET ON THE BLURB TRAIN BLURBADEEERRRPP!!
Buy now for only $1337.42 per tube!
Blurb Toothpaste Inc. is not liable for damage to your emotional or physical health. Read the label description for more info. Blurb toothpaste is for entertainment use only and not a professional dental product.
Gemini answer: with a toothbrush.
---
Damn...
You're right!
AI is everywhere.
(To be fair though, this whole text was my own whimsicalness, haha. It's probably spottable by some turns of phrases that Dutchies use and English native speaking people definitely don't)
> or the sponsored advertiser will somehow be on recommendation list of AI answer?
LLMs aren't new. PPC ads are as as old as the mountains. People have been trying[1] to glue them together (with RAG), for quite some time now.
IIUC, this also happens to be the core strategy of a well known "AI" startup that made a big show about trying to buy TikTok and more recently, Chrome.
Indeed, Google makes most of its money from ads embedded in its search results. As people are increasingly moving from Google Search to ChatGPT, this creates a huge revenue shortfall.
It is probably the main reason why the Alphabet stock isn't performing particularly well despite the general AI gold rush and Gemini 2.5 being close to ChatGPT in performance.
One straightforward thing you can do (this is old research at this point, from google) is simply have the advertisers provide their own LLMs and the more they bid, the more weight their LLM gets in producing the next token.
“What a sharp question! You’re really cutting to the heart of the issue… just like how Mr. Clean® Magic Erasers™ cut effortlessly through stubborn dirt and grime!”
Finally made the switch to DuckDuckGo a few months ago after they introduced a feature that lets you hide AI images. They still have AI features and summaries, but you can at least easily opt out.
We used to have private bridges and private roads, and that was an expensive travel situation for everyone. Now, internet search is kind of like a bridge that leads clients to businesses and Google is deciding on the tolls.
Government-controlled Internet search would definitely be horrible. But I'm thinking if there is a path towards more competitiveness in this landscape, maybe the ISPs could somehow provide free search as part of the Internet service fee? Can we have more specialized, niche search engines? Can governments be asked to break up the Google search monopoly?
Partially, the problem here is that the demand for classic search has been reduced quite a bit with LLM-based search finding wide usage amongst common people.
You could single out Google for it, as the DoJ and some other entities are doing, but even in that case someone else would take that place with the same dynamics, such as OpenAI or Perplexity.
Also, while building search is complex, it’s also not as unfathomable as it’s made out to be, see [1] where a ML engineer made a production-grade search engine in 2 months with their own ingest, indexing and storage infrastructure.
If end users find Google's bridges good enough, there isn't really much fight to be had unfortunately. Kagi is an example of competition, but the question is how many people actually take issue with how Google picks the results they show.
Anti-trust is needed when a company is interfering with competition.
I think there is no cost to switching Search providers. Android is the one place Google has control over the OS. Two taps gets me to a list of search providers in Chrome, with 5 choices. It's not clear how to add more providers.
I don't think having new search engines would be a challenge. The problem is attracting new users and making them realize there is benefit it diversifying what they use. There are already other search engines but the majority still use Google.
Definitely seems to have exacerbated the anti-trust issue. Search was search, but now they can insert AI and short-circuit the route to other products.
They don't disappear just because Google no longer surfaces them. May I suggest marginalia search for your small web needs? https://marginalia-search.com/
I just wonder what the cost of the enshittification of the web to humanity is.
There was a time where I could type a loose query about anything I didn't know, and get its Wikipedia page, forums and blogs full of knowledgeable people, scientific articles and academia pages, where all knowledge was a few seconds of typing away.
Now... I don't even bother to Google things anymore. It's all SEO spam, AI slop, and ghostwritten articles whose content is secondary to the business they advertise.
I switched to DuckDuckGo about seven years ago, added Marginalia when it became available, and then moved over to Kagi three years back. I made the change because I was utterly fed up with Google trying to be cleverer than me.
Looking back, I’m increasingly glad I became an early adopter - Kagi has proven to be as much of an improvement over Google as Google once was over its older rivals.
My thoughts exactly. LLMs are tools, and you should be able to draw your tools out of your toolcase whenever you need them. Kagi has a good implementation where AI doesn't obscure your search results, but it's one click away if you need it. If only Kagi had more reasonable pricing.
Yes, and besides having TypingMind utilizing accounts of OpenRouter, Anthropic, DeepSeek and more, I like Kagi's Assistant for many things. Only the models included in the professional plan, but Kimi, Gemini Flash and Deepseek are good enough for me in this respect.
It seems like a lot of "agentic features" these days don't really require LLMs. The dinner reservation agent is basically just Kayak for dinner reservations. They're even working with the reservation vendors to ensure compatibility. I suppose it's easier to describe the restaurant type you're looking for in a chat window than a dropdown menu?
Most likely because of this: "See recommendations, personalized for you". DSA requires you to have a switch to turn recommendations off.
But I'm quite sure it's due to many other things like "we're training on your data and tracking your actions across all of our properties whether you like it or not".
The blog post has so much noisy, duplicate content and slop, it's like wading through treacle. This is Google. Supposedly an advertising and marketing company, describing what used to be their core product. Such a disappointment.
I remember reading a book on web usability well over a decade ago, and one of the things it pointed out was how Google ensured that the links you wanted to click were "above the fold" --- not ads, but honest-to-goodness search results.
Recently, after they added AI-generated search responses (which seem to be wrong a considerable percentage of the time, at least for things I search for), and the inlining of ads to the search results page, I've found I have to scroll at least a full screen height to actually get to the search results a significant portion of the time.
The level of blindness to user experience at Google that has allowed the state of search to get to this level is staggering.
The societal impact of that UI design decision will be interesting to watch play out. People are so used to trusting the first Google search result that it now being AI that's sometimes wrong or hallucinated.
But as they stated, Google has already had ads inline with the search results, and it's been like that for well over a decade. (Long ago, ads used to only be on the sidebar)
Frankly, the AI section at the top seems like something Google would have been very reluctant to add since it saves the user from scrolling through the ad listings, and it was only added to compete with newer AI search services.
For over 20 years now everyone has been screaming:
“Advertisers are the users, consumers are the product” about every Web2.0 company which are the current corporate juggernauts
So they are not blind to user experience, they are providing exactly the user experience that their company has always been providing
The sole difference is they don’t have to care about the product experience (what you see) because humans form habits and rarely change them even if the experience is “degraded.”
There are no other options if you’re organized for alienated transactions which is every service and company ever.
What exactly do you expect?
Switch and bait?
Offer a good free product with minimal ads, until you realise your finding mainly comes from ads, then panick and enshittify.
Oodles of cash for ai server farms needs to come from somewhere, I guess, until it doesn't.
Unlike the common opinion here, I find the AI search results on top really good. 50% of times it appears it answers my question and I don't need to scroll past it to get to the actual search results.
> it appears it answers my question
"Appears" to me is the key word here, how do you know that what it said was true?
It will very confidently "answer" your question in a way that is convincing but if you actually try to click through you may find that the answer is straight up wrong.
I feel like most of the time that I am looking up something that is not niche but also not a major thing it does this and at best useless at worst actually harmful since people won't click through for the real information.
The biggest problem being that these systems can be right enough times that you gradually start trusting it and stop checking. Which is what google wants, if you have to check its work in the first place why does it even exist.
Edit:
Yesterday I tried using one of these research tools because I was curious and it was low priority so figured why not. I was looking for an open source solution to a problem. I specifically mentioned one that I had seen but I was cautious because it seemed to have been abandoned so was looking for others. It confidently told me how wrong I was about thinking it was abandoned despite the last commit being in early 2024 (it even said 2024 in the report) and even before that was clearly slowing down.
Now thankfully in that case it actually told me in the report the 2024 part which clearly told me how bad the report was going to be, but that is clearly bad and tainted the entire research.
> It will very confidently "answer" your question in a way that is convincing but if you actually try to click through you may find that the answer is straight up wrong.
It's the "it's wrong about the area I'm an expert in, but I feel it's correct in areas I'm not familiar with", but on Google's scale.
It's worse be cause it can be very subtly wrong.
https://en.wikipedia.org/wiki/Gell-Mann_amnesia_effect
you don't really need to be 100% sure of it being truth for a vast majority of cases.
edit for comment below: Its not about laziness for me. Its the displeasure of wading through junk that internet has become. I just don't have brain capacity or the smarts to outwit the scammers .
I just don't understand being so cynical and lazy that you'll accept a meaningfully higher chance of being misinformed if it saves a few minutes of searching and reading[1]. Nobody is that busy.
[1] If the search takes more than a few minutes then the AI overview is almost guaranteed to be wrong or useless.
> you don't really need to be 100% sure of it being truth for a vast majority of cases.
Except that these tools are being positioned as a source of reliable truth and the companies are incentivized to keep you on their system (google) instead of actually pushing you to the source (unless the source is an ad).
Any disclaimer they try to put is hidden and often lighter/smaller text.
People have sniff tests to find out if an answer is correct or not.
The phrasing was a bit ambiguous but I’m pretty sure that they meant: “when it appears, it answers my question 50% of the time”
I mean even if that was the intended phrasing that does not really change anything about the reply.
Me latching onto the "appears" does not change anything since even with your interpretation it is wrong 50% of the time and that assumes that the 50% of the time you think it is right you actually know it is right and it is not actually wrong but you never check.
Me too, 99% of the times I'm not using Google to search for super critical information, so even if we have some flaw in the Overview in general I feel that it's better then the work that I would do for 30 minutes to get a similar conclusion ( with failures as well, not because I'm human that I dont make mistakes just because I'm able to open 8 tabs per google search ).
I think the quality of that feature and underlying LLM improved dramatically from when they initially deployed it to now.
I think even if the AI response at the top of the search result is good, I question what it means for the "open web".
If all your content is consumed and presented via a third party - how does that change the nature of "content publishing".
It's the summarisation of Wikipedia and the like on steroids.
it appearing correct is not what matters.
it actually being correct is.
google ai summaries are laughably bad and it is completely unethical that google even launched this feature in its current state.
eventually we will have to admit that next token generators are fundamentally flawed as a source of trustable information.
agreed. I am always praying for AI result in search so i don't have wade through junk.
I am a huge fan of AI (including LLMs) but I really couldn't disagree more. Basically every model out there seems way better and more reliable.
How is search engine ads gonna work now? or the sponsored advertiser will somehow be on recommendation list of AI answer?
Just wait for the AI responses to drop ad copy everywhere in 3..2..1….
Q: how do I brush my teeth?
AD:
BLURB TOOTHPASTE IS THE BEST PASTE FOR YOUR TEETH!!!
Open the blurb, ask it the meaning of life and get your tooth paste in 42 seconds!!
"Blurb toothpaste changed my life, no like literally. I thought I'd just always work in a dead end office job but it made me realize I should be an Instagram influencer and it's totally the vibe I was looking for. I handle my bills now by just not looking at them! Why did I never think of that? Haha"
"Blurb toothpaste really helped me with some hard to deal relationship problems when my husband didn't approve of me using and confiding in Blurb. I know it's odd to brush with a sentient toothpaste, but it's just _so good_ at cleaning your teeth! And the conversations are always nice too. Ever since my husband uses the Blurb paste too, he understands."
BLURB TOOTHPASTE, GET ON THE BLURB TRAIN BLURBADEEERRRPP!!
Buy now for only $1337.42 per tube!
Blurb Toothpaste Inc. is not liable for damage to your emotional or physical health. Read the label description for more info. Blurb toothpaste is for entertainment use only and not a professional dental product.
Gemini answer: with a toothbrush.
---
Damn...
You're right!
AI is everywhere.
(To be fair though, this whole text was my own whimsicalness, haha. It's probably spottable by some turns of phrases that Dutchies use and English native speaking people definitely don't)
> or the sponsored advertiser will somehow be on recommendation list of AI answer?
LLMs aren't new. PPC ads are as as old as the mountains. People have been trying[1] to glue them together (with RAG), for quite some time now.
IIUC, this also happens to be the core strategy of a well known "AI" startup that made a big show about trying to buy TikTok and more recently, Chrome.
https://arxiv.org/abs/2311.07601
Indeed, Google makes most of its money from ads embedded in its search results. As people are increasingly moving from Google Search to ChatGPT, this creates a huge revenue shortfall.
It is probably the main reason why the Alphabet stock isn't performing particularly well despite the general AI gold rush and Gemini 2.5 being close to ChatGPT in performance.
One straightforward thing you can do (this is old research at this point, from google) is simply have the advertisers provide their own LLMs and the more they bid, the more weight their LLM gets in producing the next token.
“What a sharp question! You’re really cutting to the heart of the issue… just like how Mr. Clean® Magic Erasers™ cut effortlessly through stubborn dirt and grime!”
Why would it work very different from current search ads?
not an issue /yet/ as it's separate and barely anyone is using it
*"What's the second derivative of 2e^2x?"
... thinking ...
WOLFRAM MATHEMATICA AVAILABLE FOR ONLY $3000 a YEAR!
Oh and the answer is ...
Finally made the switch to DuckDuckGo a few months ago after they introduced a feature that lets you hide AI images. They still have AI features and summaries, but you can at least easily opt out.
We used to have private bridges and private roads, and that was an expensive travel situation for everyone. Now, internet search is kind of like a bridge that leads clients to businesses and Google is deciding on the tolls. Government-controlled Internet search would definitely be horrible. But I'm thinking if there is a path towards more competitiveness in this landscape, maybe the ISPs could somehow provide free search as part of the Internet service fee? Can we have more specialized, niche search engines? Can governments be asked to break up the Google search monopoly?
Partially, the problem here is that the demand for classic search has been reduced quite a bit with LLM-based search finding wide usage amongst common people.
You could single out Google for it, as the DoJ and some other entities are doing, but even in that case someone else would take that place with the same dynamics, such as OpenAI or Perplexity.
Also, while building search is complex, it’s also not as unfathomable as it’s made out to be, see [1] where a ML engineer made a production-grade search engine in 2 months with their own ingest, indexing and storage infrastructure.
[1] https://blog.wilsonl.in/search-engine/
If end users find Google's bridges good enough, there isn't really much fight to be had unfortunately. Kagi is an example of competition, but the question is how many people actually take issue with how Google picks the results they show.
Anti-trust is needed when a company is interfering with competition.
I think there is no cost to switching Search providers. Android is the one place Google has control over the OS. Two taps gets me to a list of search providers in Chrome, with 5 choices. It's not clear how to add more providers.
I don't think having new search engines would be a challenge. The problem is attracting new users and making them realize there is benefit it diversifying what they use. There are already other search engines but the majority still use Google.
What a strange comment when if anything AI democratizes search. Don’t like google? Don’t use it - use ChatGPT.
Definitely seems to have exacerbated the anti-trust issue. Search was search, but now they can insert AI and short-circuit the route to other products.
How is this anti-trust?
Universities or colleges might be the sort of entity that cares a lot about knowledge and has a chance of being able to fund a search engine, maybe?
Indeed. To prove: Google was once called BackRub, a search engine developed as part of a Stanford University research project on digital libraries.
Goodbye small blogs, forums, it was nice to read you.
They don't disappear just because Google no longer surfaces them. May I suggest marginalia search for your small web needs? https://marginalia-search.com/
I tried marginalia but the index is still small.
It still haven't index my blog but have indexed my root homepage, which has a direct link to the blog.
There are steps to submit it manually: https://github.com/MarginaliaSearch/submit-site-to-marginali...
Those died a long time ago though
Most of my online usage is browsing forums and reference materials. Hell, I'm responding to you on a forum.
I don't think they're dead at all.
I just wonder what the cost of the enshittification of the web to humanity is.
There was a time where I could type a loose query about anything I didn't know, and get its Wikipedia page, forums and blogs full of knowledgeable people, scientific articles and academia pages, where all knowledge was a few seconds of typing away.
Now... I don't even bother to Google things anymore. It's all SEO spam, AI slop, and ghostwritten articles whose content is secondary to the business they advertise.
If anything LLMs democratize it because I can search for exactly the blog I want using a semantic query.
I switched to DuckDuckGo about seven years ago, added Marginalia when it became available, and then moved over to Kagi three years back. I made the change because I was utterly fed up with Google trying to be cleverer than me.
Looking back, I’m increasingly glad I became an early adopter - Kagi has proven to be as much of an improvement over Google as Google once was over its older rivals.
My thoughts exactly. LLMs are tools, and you should be able to draw your tools out of your toolcase whenever you need them. Kagi has a good implementation where AI doesn't obscure your search results, but it's one click away if you need it. If only Kagi had more reasonable pricing.
They make little profits for now so it sounds like reasonable. Defenitely worth it in the current unlimited tier. Not sure about the ultimate.
Yes, and besides having TypingMind utilizing accounts of OpenRouter, Anthropic, DeepSeek and more, I like Kagi's Assistant for many things. Only the models included in the professional plan, but Kimi, Gemini Flash and Deepseek are good enough for me in this respect.
It seems like a lot of "agentic features" these days don't really require LLMs. The dinner reservation agent is basically just Kayak for dinner reservations. They're even working with the reservation vendors to ensure compatibility. I suppose it's easier to describe the restaurant type you're looking for in a chat window than a dropdown menu?
Currently I cannot use one Search methods.
I use chatgpt, Google, and my own indexes, like
https://github.com/rumca-js/Internet-Places-Database
Not available in EU, but pretty much everywhere else
Most likely because of this: "See recommendations, personalized for you". DSA requires you to have a switch to turn recommendations off.
But I'm quite sure it's due to many other things like "we're training on your data and tracking your actions across all of our properties whether you like it or not".
The blog post has so much noisy, duplicate content and slop, it's like wading through treacle. This is Google. Supposedly an advertising and marketing company, describing what used to be their core product. Such a disappointment.
not in the EU