Google introducing AI to search results

Joined
Sep 3, 2015
Messages
811
Likes
582
Degree
2
In the coming weeks and months, we’ll make these language models available, starting with LaMDA, so that people can engage directly with them. This will help us continue to get feedback, test, and safely improve them. These models are particularly amazing for composing, constructing, and summarizing. They will become even more useful for people as they provide up-to-date more factual information.

https://www.seroundtable.com/google-pichai-confirms-chat-based-search-34849.html

I don't think this will be positive to publishers. Thoughts?
 
An excerpt from that article:

We have been preparing for this moment since early last year, and you’re going to see a lot from us in the coming few months across three big areas of opportunity; first, large models. We published extensively about LaMDA and PoN, the industry’s largest, most sophisticated model plus extensive work at DeepMind,”

Take note of the bolded words. It pretty much explains/confirms my suspicions that 2022 algo updates were more or less Google continuously testing something on the live algorithm and not so much to do with "rewarding better content".

Back to the main topic, i don't have much to say but take note of this statement:

"These models are particularly amazing for composing, constructing, and summarizing. They will become even more useful for people as they provide up-to-date more factual information."

It sounds like first phase of implementation will mostly affect certain types of informational queries (i.e what is xxx, how to do xxx, e.t.c) and featured snippets will see the biggest changes.
 
Let's see how this plays out. It depends on how its implemented.

That being said, I think informational sites will be more affected than product review sites, because from a user perspective, chatting with AI might give you answers to questions, but it won't be able to recommend the best product based on your situation or need (in most cases). Even if it did, I still think users want to look and compare on their own. Maybe that's why they have been pushing "product review, helpful content updates" so frequently. The timeline matches up with what they are say is preparation for the launch of their AI.

I don't think informational sites will die either, I am just saying they will be affected more than product review sites.

I also see a major issue with these AI tools because they are using publishers content (without their consent) to learn from and use their hard work to eventually profit from. It's kind of like what Google is already doing by showing the best answer straight in the search results (position 0) to get users to stay longer on their website and therefore dropping the CTR to those sites. They also don't give credit to those publishers. AI will not give credit to the publishers of the information it learned from either lol

What I am going to do is diversify my income through other channels. YouTube, Pinterest, etc...

I also think publishers should form a union to protect their rights and hard work (I don't know of one that exists), otherwise we will just keep getting stepped on.
 
we will just keep getting stepped on.
Getting stepped on from getting FREE traffic from the domain google.com?

Google doesn't owe anyone anything. If you want their traffic, play by their rules. That requires putting your data/content inside their databases which they'll turn around and do whatever they want with it.

Or just block Googlebot.

It's a cut throat world out here. There is nothing fair about reality, from governments, to relationships, to business.

It's sort of like religions that believe "If I do the right thing here on earth, I'll be rewarded in the AFTER LIFE - when I'm dead."

Well, maybe, or more likely you'll be dead and have no body to do anything "rewarding".

But religious thinking has crept into society, and in SEO, "If I just do correctly this way, I'll eventually be reward." Well, maybe, maybe not.

Just like the universe is indifferent - you are just a speck of stardust to it. Your governments are indifferent - you're just a number to them. Google is indifferent - you're just a number to them too.

If your website is how you feed your family, you have to do whatever it takes to get traffic and that means getting more than a single website, google.com, be send you their traffic.

We can wait and see but that's the SEO's default position, waiting till the next algorithm update. Whatever happens within A.I. is just going to be a new ranking factor/part of reality of online traffic.

Unions? If that's your option to strong-arm Google that's an victim-mentality battle.

You guys needed to diversify traffic sources A LONG TIME AGO. It's only going to get tougher.
 
Unions? If that's your option to strong-arm Google that's an victim-mentality battle.

There is nothing wrong with unions. I don't care if you label them as victim-mentality or lazy ass people or whatever you want. Unions are the reason people get paid fare wages among a lot of other benefits. Even Google/Alphabet have their own union: https://alphabetworkersunion.org/people/our-union/

No one is forcing anyone to enter a union either. It's a choice, just like deciding to take a stance on your own without any help or support.
 
There is nothing wrong with unions.
You're not Google employees. They send you free traffic from their website. You can get free traffic from other websites.

Imagine some group of people coming to you and saying, "hey, stop changing the ways you calculate the free stuff you send us."

What? That sounds crazy.

If you have a problem just block them.
 
Definitely not positive for publishers. I've been anticipating this. While everyone was worried about AI content over saturating markets, I was more worried about AI disrupting search itself. I just mentioned last week in another post about google likely implementing their own AI in place of featured snippets, and here it is now. It makes sense for Google to implement this for the benefit of users, and to prevent any competition from taking market share. And hey, how convenient, it'll keep users close to the ads. SEOs will be complaining, because thats what we do. We will get less clicks. The trend will continue. However, this is about to accelerate fast. AI answers offer infinitely more utility to the end user than featured snippets, especially as it progresses (we are only at day one).

It will absolutely capture a good amount market share of clicks. Let's say 25% of people click the first result now. With AI, maybe that drops down to 15% or 10% because now people don't need to click through, because their query is sufficiently answered.

Not being doom and gloom over here. I just see it for how it is. Adapt or die. Maybe jump into E-commerce more. Video is probably the next best opportunity for informational queries, until we can make AI avatars combined with AI scripts that automates Youtube videos. Pivot when needed. There will still definitely be a place for SEO, its not going to disrupt search completely, but certain (many) informational queries are likely to have a good amount of traffic eroded away by AI answers.

However, I would say the trend at first would be informational queries that by nature have a short answer (where a quick answer suffices). So I still think long, in depth content pieces will continue to do well. What happens in the future, if google creates like another section where it instead of creating little AI snippets, AI spits out full comprehensive blog posts? It may be a little bit of time for that to happen, if ever (google needs to be confident in the factual nature of AI output).

Only time will tell. But technological advances change things, and I predict this general AI will be a bigger change to society than the internet itself was. So buckle up, things are changing (not just talking about search).

There is a catch 22 though. There needs to be new content for the language models to stay up to date. So without publishers, AI answers would become outdated on newer content. So it will definitely be interesting to see what happens.

Other thoughts:
...Also, Google could have something that is much better than OpenAI and all the other language models. They've been working on it for a long time and are a huge company with the best engineers. I think it will shake things up in the world, I anticipate their AI tech to be really good.
...what happens when the learning data (the content) for these language models, become full of AI content? It'll be AI learning from AI
 
Last edited:
I saw someone pose this question elsewhere and I still think about it:

Is this going to disrupt things any more than "Hey, Siri / Alexa / Google Assistant" did?

Those things ended up being cool in terms of sending a text while driving or voice commands to call someone and all the speech-to-text features... but did people really embrace them for search in such a way that it was anything more than a goofy novelty?

Between "Hey Siri, what's 1 to the 999th power" so people can beatbox and "Hey, Alexa, fart for me", they didn't really affect anything that much.

Sure, you can ask Siri for facts, but does anyone really do that at such a volume that it matters at all? And will this text AI really be any different? Especially once people realize, just like they did with the voice ones, that they're simply reading crap back from the internet that's likely incorrect?

It's been fun to see people outside of the tech world start talking about "general AI" and "true intelligence" and all of that, but it's really still nothing more than a content scraper and spinner. It spins syntax instead of just adjectives and nouns, but that's not been new for a long time. Beyond that, what's it really doing that's that new?

Interestingly, the thought just came to mind that this is probably a large part of why Google started doing "passage indexing".
 
I saw someone pose this question elsewhere and I still think about it:

Is this going to disrupt things any more than "Hey, Siri / Alexa / Google Assistant" did?

Those things ended up being cool in terms of sending a text while driving or voice commands to call someone and all the speech-to-text features... but did people really embrace them for search in such a way that it was anything more than a goofy novelty?

Between "Hey Siri, what's 1 to the 999th power" so people can beatbox and "Hey, Alexa, fart for me", they didn't really affect anything that much.

Sure, you can ask Siri for facts, but does anyone really do that at such a volume that it matters at all? And will this text AI really be any different? Especially once people realize, just like they did with the voice ones, that they're simply reading crap back from the internet that's likely incorrect?

It's been fun to see people outside of the tech world start talking about "general AI" and "true intelligence" and all of that, but it's really still nothing more than a content scraper and spinner. It spins syntax instead of just adjectives and nouns, but that's not been new for a long time. Beyond that, what's it really doing that's that new?

Interestingly, the thought just came to mind that this is probably a large part of why Google started doing "passage indexing".
I'm pretty-pro AI but I think you're looking at this wrong. Asking Siri for the answer isn't super popular because it's a fucking goofy way to search, especially since most of us are not alone 100% of the time. The impact of this is going to be more similar to the impact of featured snippets in Google, which means, a lot. It directly removed a huge % of traffic for sites doing the whole "what's the weather in X" play.

But for results where you need more context, people still generally click. If you just end up getting the full answer in the SERP, clicks drop massively. So, people aren't not used Siri because it's "simply reading crap back from the internet that's likely incorrect", it's just because of the input requiring voice. You can see this to be true, because snippets dropped traffic, so people clearly feel comfortable to rely on just a tiny snippet Google gives. They'll rely on it even further when they get more detailed, nuanced replies.

Having said that, I still don't think Google will introduce this into main search for years, without links. Liability is way too high. So, they need to include the link. If they include links, they need some incentive mechanism to reward content writers to keep a flow of fresh content. So, it probably ends up looking like some sort of Youtube revenue share program if I had to spitball.
 
I saw someone pose this question elsewhere and I still think about it:

Is this going to disrupt things any more than "Hey, Siri / Alexa / Google Assistant" did?

Those things ended up being cool in terms of sending a text while driving or voice commands to call someone and all the speech-to-text features... but did people really embrace them for search in such a way that it was anything more than a goofy novelty?

Between "Hey Siri, what's 1 to the 999th power" so people can beatbox and "Hey, Alexa, fart for me", they didn't really affect anything that much.

Sure, you can ask Siri for facts, but does anyone really do that at such a volume that it matters at all? And will this text AI really be any different? Especially once people realize, just like they did with the voice ones, that they're simply reading crap back from the internet that's likely incorrect?

It's been fun to see people outside of the tech world start talking about "general AI" and "true intelligence" and all of that, but it's really still nothing more than a content scraper and spinner. It spins syntax instead of just adjectives and nouns, but that's not been new for a long time. Beyond that, what's it really doing that's that new?

Interestingly, the thought just came to mind that this is probably a large part of why Google started doing "passage indexing".

exactly my opinion

it's all cool and stuff until people find out it's a nuisance to actually work with and then it isn't cool anymore
 
Screenshot_2023_02_03_at_17.38.29.png

FoSR19RWAAAo0u5


this was in the Bing UI for a little bit. This is how they are doing it. No link etc.

Seems to be removed already, was only live for a few hours. It's just straight ChatGPT but with a modifier to not reply if not confident
 
Not being doom and gloom over here. I just see it for how it is. Adapt or die. Maybe jump into E-commerce more. Video is probably the next best opportunity for informational queries, until we can make AI avatars combined with AI scripts that automates Youtube videos. Pivot when needed. There will still definitely be a place for SEO, its not going to disrupt search completely, but certain (many) informational queries are likely to have a good amount of traffic eroded away by AI answers.
This already happening at full scale with AI voice and AI avatars.

Ive seen people copying the text scripts of the best performing videos on youtube in different niches, taking it to Chatgpt for rewriting it and making a AI voice speak together with a AI avatar and they are pushing out 3-5 videos a day and lots of people are seeing huge success after 30 days even.

But I belive that people will notice what is AI produced after a while and people will get tired of it and lean towards real creators with experience.
 
If you watch how non-opportunists are playing with ChatGPT and every other AI remixer app, it's being treated as a toy. I'm not trying to downplay it, especially not out some magical thinking that a hope that it won't catch on can actually influence the outcome, which I'm seeing a lot of out there, straight up coping methods.

I'm just not seeing the real deal utility of it, much like @CCarter harps on about, rightfully so, concerning cryptocurrencies. It's not really solving a problem. It's just a new way to search the internet, and people in general don't even care about that. They care about it as a new toy to ask funny or agitating questions and soon they'll tire of it and throw it in the toy box in the attic eventually.

It's also being politicized and gimped to push certain narratives and restrict the flow of certain information, which is going to reduce the trust levels just like with Siri and Alexa fetching bad information. That stuff matters far more than the narrative allows anyone to think about, since even that information gets suppressed and poo-poo'd and lied about. No amount of telling people to disbelieve what's sitting right in front of them will ever work, which is why it next, like always, moves into the punishment phase. All of this will play out in the "AI field", as it is everywhere else.

That's not to say it won't capture (or create new) attention and revenue and pull it away from others (SEO's). It's just not the game changer it's being made out to be, in my opinion. There's a big lawsuit gearing up regarding the AI Graphics Generators, because all they're doing is "spinning" other people's artwork, which may not remain legal. Nothing these things are doing reflect actual intelligence, yet. It's like sampling in music. It's legal when you pay royalties and get permission, neither of which any of these bots do.

Ive seen people copying the text scripts of the best performing videos on youtube in different niches, taking it to Chatgpt for rewriting it and making a AI voice speak together with a AI avatar and they are pushing out 3-5 videos a day and lots of people are seeing huge success after 30 days even.
They're in for a surprise. I ton of that stuff got demonetized with the new ruleset that rolled out. Youtube has rules about simply repurposing content (even the script) from another place and then auto-generating videos out of it. It's all fair game with "fair use" but Youtube's own line in the sand is to not fill up their platform with auto-gen garbage.

The quote below is from here: https://support.google.com/youtube/answer/1311392?hl=en

Content that violates this guideline

When a channel’s content consists of similar content, it can frustrate viewers who come to YouTube for appealing and interesting videos. [...] In other words, your channel shouldn’t consist of content that’s automatically created or produced using a basic template.

Examples of what’s not allowed to monetize (this list is not exhaustive):
  • Content that exclusively features readings of other materials you did not originally create, like text from websites or news feeds [...]
  • Similar repetitive content, or mindless content with low educational value, commentary, or narrative
  • Templated, mass-produced, or programmatically generated content
  • Image slideshows or scrolling text with minimal or no narrative, commentary, or educational value

^ This stance is the kind of stance I don't think Google can or will take on AI generated content, though their guidelines do state that mass-generated AI content explicitly for the purpose of ranking is a no-no. They'll de-rank it, where as Youtube will do the same. They won't delete it, but they're not going to allow monetization or wasted bandwidth.

So, people aren't not used Siri because it's "simply reading crap back from the internet that's likely incorrect", it's just because of the input requiring voice.
That's not quite what I said or what I meant. It is a piece of the puzzle. It didn't survive the novelty exploration phase because it didn't bring anything new to the table, just another way to interact with what already existed.

A counter argument to the "it's because it requires a voice input" in public would be that people have no problem having conversations in public instead of writing down messages on paper and passing them back and forth. They have no problem talking on the phone in public. And they have no problem blasting music out of little bluetooth boomboxes, either.

The voice input is the feature that makes it attractive, in public or private. The problem is, even in private people aren't interacting with the internet with it. They might use it to interact with the internet-of-things (IoT) or some walled garden apps (Siri, play Ying Yang Twins, turn the volume down a bit) but not the internet as we know in terms of text and video content.
 

Google Bard is now being tested!​

Bard, the new Google AI Chat module is being soft-launched to "trusted testers", and this is what they've shared to show what it looks like:

2BhqC7N.png


6n5V4QL.png

Barry Schwartz asked Google about how they plan to give links and attributes:
I asked Google how it plans to provide attribution and/or citations for the answers Bard provides to searchers. Google did not have an immediate comment at the time this was published.

Things are moving fast once Bing stepped up to the plate.
 
There's a big lawsuit gearing up regarding the AI Graphics Generators, because all they're doing is "spinning" other people's artwork, which may not remain legal.
I can see the EU having a field day with Google on this. That might help protect publisher traffic.

Google Bard is now being tested!​

Bard, the new Google AI Chat module is being soft-launched to "trusted testers", and this is what they've shared to show what it looks like
Wow that is pretty scary to be honest. Didn't expect 3 full paragraphs. Informational sites of any sort would be dead for all intents and purposes.
 
Wow that is pretty scary to be honest. Didn't expect 3 full paragraphs. Informational sites of any sort would be dead for all intents and purposes.
in my opinion, the example they gave for the SERP widget is a completely useless non-answer and nothing more than an "improved" (for google's bottom line that is) version of the snippet

i cant be the only one who feels that it's closer to "we just had to find a reason to remove another outgoing link from the top of the SERP" than "the AI revolution is here"

realistically speaking, nobody who is looking to pick up a new instrument and invest countless hours on it will make their decision by reading something that practically says "some say piano is easier to learn, but others say guitar is easier to learn", which sounds like something you would read on an income school blog at best

in its current form, i simply don't see how this format can produce a satisfying answer for any informational queries where detail is required (unless they just take the entire screen space to go into detail like the chatbot does, which i guess they are avoiding for certain reasons)

the best it can do is take traffic from news sites by directly plagiarizing basic news stuff or answer things like "how long to cook chicken in the oven", which are usually answered by snippets or some other form of SERP feature anyway
 

You've been calling out Google's slow transition from being a "Search Engine" to an "Answer Engine" for a few years now, and this is as "Answer Engine" as it gets.

I was wondering how they would implement their Ai Chatbot feature. To see it baked directly into the search bar like this is pretty wild. Instead of showing a snippet from a reputable resource, with link attributions, they're just scrapping and spinning to create an "AI" answer. I'd imagine the author of the article who researched the figures for this answer would be unhappy with this scrape-and-spin game.

Perhaps the links under "read more" are the source sites, but how long before Google places ads directly under these Chatbot answers?

Perhaps they'll only use this Ai for ultra-longtail queries requiring more of an explanation as a response - who knows, but they're moving fast.

I really can't stand all these Ai writers - makes outsourcing content tricky, and to now see Google's plan to return Ai answers directly in the search field could change the face of search. I'm hoping this is all a fad that dies off and Google will get back to being a search engine.
 
in my opinion, the example they gave for the SERP widget is a completely useless non-answer and nothing more than an "improved" (for google's bottom line that is) version of the snippet

i cant be the only one who feels that it's closer to "we just had to find a reason to remove another outgoing link from the top of the SERP" than "the AI revolution is here"

realistically speaking, nobody who is looking to pick up a new instrument and invest countless hours on it will make their decision by reading something that practically says "some say piano is easier to learn, but others say guitar is easier to learn", which sounds like something you would read on an income school blog at best

in its current form, i simply don't see how this format can produce a satisfying answer for any informational queries where detail is required (unless they just take the entire screen space to go into detail like the chatbot does, which i guess they are avoiding for certain reasons)

the best it can do is take traffic from news sites by directly plagiarizing basic news stuff or answer things like "how long to cook chicken in the oven", which are usually answered by snippets or some other form of SERP feature anyway
It's essentially dynamic featured snippets, which more detail because it consumes information across multiple sites.

If featured snippets took 5% of clicks away on 15% of SERPs (guessing in 2023), I'd lick my finger, put it in the wind, and say that this takes away about 10% of clicks from 20-30% of SERPs. That's a major difference. You're talking about potentially 4x the impact of featured snippets.

People WILL rely on the information from the AI. It's not even up for debate. 5% of people are already relying on featured snippets, which are far worse. This will reduce traffic from Google to sites, no doubt.

On the flip side, someone will win here. If your site has, as Ted Kubaitis coined it, "high information payload" and the AI relies on it a lot for the output, you might find yourself under the AI Snippet, when you were ranking #23 before. So, for sites where the content quality is strong, but perhaps the page doesn't rank well under the traditional Google model, you could end up happy with this change.

For sites that rank highly already, this is without a doubt a negative change.

I agree that the EU is going to come after this hard, this is the kind of shit that they get wet thinking about. Conservatives in the US will jump on this as soon as a single site on their side of the aisle claims to have been "DESTROYED" by Bard. But, Google will likely float away nearly unscathed and just be forced to do better attribution to the site where they get most of the "informational payload" from, the same as what happened with Featured Snippets years ago.
 
What if I write a super specific article on something. With content and calculations no one else has done. Google would then take that content, spin it, and spit it back out as AI with no attribution?

That's not possible. And the answer isn't to block Google. You shouldn't need to actively block someone from stealing your content. Copyright law and theft doesn't work like that.
 
One thing people forget is that this will vastly increase the variety of questions people will type into Google. So while you may receive fewer clicks per search, this may (!) be offset by an increasing number of new queries.

Also, I'm surprised to see that no one has pointed out the obvious: it is absolutely not in Google's interest to answer every question you have. They make money from both search and display ads (even sites on MV and AT since they run through Google's Ad Exchange), so there's a great incentive to push users to the actual website.

We'll see how this all plays out. I can certainly imagine a world where nobody gives a crap about Bing's AI features and doesn't adopt it, which then prompts Google to dial back its efforts as well.
 
The most ironic part of all this is Google began categorizing the use of AI as spam. Yet, they are implementing AI for the same purpose.

How Google implemented this is exactly the way I envisioned it. It’s going to featured snippets on steroids.

By the way, I am pro this technology, err “AI”, and very excited about it. Yes, it will suck for SEO and some queries. It will have a far greater impact than the old featured snippets in the effect it has on clicks. If you don’t see this, I almost think there must be some bias or you are too skeptical of the technology or tying it too much to past experiences. Make no mistake, this is different. There are many smart people on this forum that I respect, but I’m surprised more people aren’t grasping the magnitude of this…

I’m not saying it’s not an SEO killer. Not by a long shot. But I am saying it will 100% take significantly more clicks away than featured snippets did when those were introduced. I’m not typically one to fall prey to hype. But guys, this is not a hype thing. It may have hype following its wake, but it’s not based on a speculative hype; there is utility. Sure, the term AI is full of hype. Sure, it’s not true AI. But there is a real utility to this technology. And, in my mind, the utility goes way beyond content creation or search. Ultimately, it’s in a different league and context of voice assistants.

1. It is not a true AI. It’s not conscious. It’s not truly intelligent. The word AI is one of hype (one day we will have true, intelligent, sentient AI, and that will really change things)
2. It is a language model. It is phenomenally impressive. The technological leap between the old content spinners and these language models like ChatGPT is immense. The implications are also immense. Eventually, it will be indistinguishable from human content, at least for detection purposes. You always need the human element (i.e the prompts to build good content pieces, write content, etc.)
3. It will be a benefit to early adopters. It will hurt some businesses in the transition stage, it will start being trained for specific industries. Eventually, people will be saying: “THEY TOOK ‘ER JOBS!”, as machine learning, language models, etc. outpace the creation of new tech jobs... Then it will offer society a productivity benefit in the long term. From an economic standpoint, capital owners always benefit more from productivity increases.
4. Assistants: Google scrapes sites to place a snippet of the first page. It is sometimes limited. It only gives a glimpse of an answer. Sometimes it suffices. Sometimes there isn’t any good data for a snippet to answer a question.
5. ChatGPT: Infinitely more powerful. It will summarize the answer to nearly ALL search queries, whether on the infinitely variable nature of long tail keywords or on a more common search. It will have Wikipedia level trustworthiness (or less) to start but it will become more accurate as it evolves.
6. Featured snippets have taken click-share away from website owners. This trend will 100% continue and multiply, with a much more useful tool people can use to get answers.
7. AI Art: There is a good argument on either side. We are in uncharted waters. The issue at hand is also a grey area. Only the courts will decide, so time will tell. I have a feeling there is a stronger argument that AI art is transformative work, which would make it eligible for fair use protection.
8. I think these language models are going to be really powerful for learning and teaching. Virtual mentors. It will make learning much more efficient. It’s a faster ‘bicycle for the mind’.

ChatGPT vs Assistants:
There is no comparison, this general AI language model will be far more disruptive than voice assistants like Siri, Alexa, or Google Assistant. This is an apples-to-oranges comparison. After spending many hours using ChatGPT, I can confidently say it is more of a comparison of t-ball to the major leagues. Siri/google offers the convenience of voice, but it lacks the granularity, speed, and versatility of interacting with a search engine by typing on a computer or phone keyboard. Thus the “novelty”, widespread adoption, but very limited usage. So it never caught on, beyond casual, convenient, quick web surfing via voice.

This “AI” will be integrated into THE SEARCH BAR. It’s where people are going to sit down at the computer or on their phone and do some searching. Not just “I’m doing dishes and going to ask a quick question to my google speaker”.
 
Last edited:
Im more worried about future google updates than google adding this summarizer… they already have this in the form of featured snippets - if anything the source will likely be mentioned the same way phind. com has - allowing more clicks to those who deserve it by linking out to the source.

People grew up and learned to search google the way its used now. A summarizer isn’t going to magically destroy everything that was built. Also, google and bing needs publishers to continue to make content and be rewarded so they cant feed their new shiny toy “bard/chatgpt”. I havent seen @Ryuzaki and @CCarter carters opinion on this, but i’d imagine they have a somewhat similar viewpoint, i would think.
 
Look at your GSC queries to see what's performing well.

Now punch that exact query into Bing or Bard when it's available. Is AI capable of producing something for search queries like programmatic SEO, SEO job boards, or starting a porn site (just examples from my site)?

Or is this designed for those millions or billions of dumbass searches Auntie Marg does because she doesn't know how to use a search engine? Please tell me how long to put my yellow sponge cake in the oven for thank you, or I have this yellowish green goo coming out of my privates what's wrong with me? Those search queries that don't make it to GSC because they're one-off searches.
 
Also, I'm surprised to see that no one has pointed out the obvious: it is absolutely not in Google's interest to answer every question you have. They make money from both search and display ads (even sites on MV and AT since they run through Google's Ad Exchange), so there's a great incentive to push users to the actual website.
This is an interesting observation. 57% of Google's revenue is from search ads, so for sure, Google will not just kill this revenue stream.

Then the question is, how valuable could it be if the brand becomes the recommended solution via this Google Bard tool? You get your brand mentioned with all the excellent features, and there are no competitors of yours visible around.

While writing this, I wondered if Google is in the business of search or ads and if the search is just the means to be in the business of ads.
 
Back