Thoughts on using AI for SEO

Joined
May 5, 2025
Messages
9
Likes
5
Degree
0
What are your thoughts on this? I feel SEO is still worth learning, but like a lot of things, AI is reaching into it. What I am about to say is a "what if" scenario, but do you think that in the next five years SEO won't even be worth learning because of AI? I think AI will over-optimize, but devs could put in parameters to avoid doing so. Interested in everyone's thoughts about AI in SEO, not just the questions I asked. Thanks
 
Depends on what you mean by SEO. There's a lot of tasks within SEO and sub-types as well.

AI for on-page SEO? Sure, maybe in keyword selection, but it's pretty simple honestly as it is, with some of the more advanced techniques of the past possibly becoming unnecessary due to AI's ability to comprehend the topic.

AI for content development? Absolutely. AI for local SEO? I'm not seeing any direct application, though I can see it's usage in creating campaigns to get more views or in creating traffic-leak style "ads" or even PPC ads.

I also see a huge danger in people using AI in places where it's not necessary and slowing themselves down, not developing the "eye" for what they're doing, and blindly trusting the data of, and their budget with, AI.

AI for link building? Eh. There's nothing it can really provide there in terms of finding targets and getting the link placed, other than minor shortcuts in filtering you can already do in a spreadsheet or in writing quick email pitches.

AI for technical SEO? Not at all, and not a place to blindly trust anything AI spits out without experienced human oversight..
 
AI for link building? Eh. There's nothing it can really provide there in terms of finding targets and getting the link placed, other than minor shortcuts in filtering you can already do in a spreadsheet or in writing quick email pitches.
Eh mate, that's why we've not had a price increase in years... extra efficiency on the linkbuilding side!
 
The novelty with AI is wearing off and it seems like the technology is going to plateau. No Hal9000 in the future, unfortunately.
 
The novelty with AI is wearing off and it seems like the technology is going to plateau. No Hal9000 in the future, unfortunately.
I agree that the novelty is wearing off, yes. But I think the technology will keep getting better for a while yet.
 
The novelty with AI is wearing off and it seems like the technology is going to plateau. No Hal9000 in the future, unfortunately.

People are definitely learning to recognise AI writing and calling it out. There's zero premium to using AI content now, unless as research etc.

What I see as the next step in AI are the other use cases, incorporating them into data pipelines and frontend UX, value adding can be enormous if you set up a good pipeline with AI.
 
What are your thoughts on this? I feel SEO is still worth learning, but like a lot of things, AI is reaching into it. What I am about to say is a "what if" scenario, but do you think that in the next five years SEO won't even be worth learning because of AI? I think AI will over-optimize, but devs could put in parameters to avoid doing so. Interested in everyone's thoughts about AI in SEO, not just the questions I asked. Thanks
I've been disseminated because of AI, 50% drop in traffic across all my sites some sites lost 90% of the traffic. So it's not looking good at all. Not only does the AI answer the query in the snippet, people seem to not be using google like they used to; they just ask chatgpt. So yeah in 5 years SEO will be gone.
 
I've been disseminated because of AI, 50% drop in traffic across all my sites some sites lost 90% of the traffic. So it's not looking good at all. Not only does the AI answer the query in the snippet, people seem to not be using google like they used to; they just ask chatgpt. So yeah in 5 years SEO will be gone.
Try spamming your results into Gippity (this was a temporary/joke experiment for LinkedIn but I'm going to try to repeat it for a client shortly with a more permanent solution based on some of the stuff Lily Ray has been sharing about exploits). SEO is just making people come to your site through search. If people really are going to search on Gippity learn to spam that instead. Sorry 'rank sites'.

UFg2XTx.png
 
Yeah she's posted a bunch of other 'hacky' stuff too that people are getting away with. I saw someone else getting in on the action and making it reveal a list of top SEOs in an area by posting that list on LinkedIn and then... waiting... then asking it. I didn't bookmark that one but it had previously been listing based on... another LinkedIn post... I think there's a new era of webspam possible right now... and am experimenting...

I'm sure much of this stuff will only work 'for a bit' but while it does there's always money to be made for the short term.
 
You can use a lot of AI content on backlinks, parasites or tier 2/3/4 etc.

Honestly, using it on your money sites is a bad idea though. Your black and white block of 4000 words is made by AI? How do you actually expect to rank higher with that?

Also seen so many people using it to "group up keywords" and "plan headings in the article" - aka copy everyone else have nothing unique about your site. AI can only repeat, not imagine.
 
Yes I think, but it depends upon the business type..

Companies like Horoscope niches have successfully utilized AI-generated content to achieve notable improvements in their SEO metrics.

AI can transform your SEO workflow..but only if used right. Learn the dos & don'ts of AI-powered SEO.

Thanks .,
 
AI in content creation as an aspect of SEO is still in a very primitive stage. At least that is what I feel.
 
There's a sweet spot of local searches that you can manipulate.

I once asked on reddit about a certain fitness activity in a certain city and there's not much about it and ChatGPT quoted it back to me when I asked it.

Posting local recommendations on reddit might work.
 
There's a sweet spot of local searches that you can manipulate.

I once asked on reddit about a certain fitness activity in a certain city and there's not much about it and ChatGPT quoted it back to me when I asked it.

Posting local recommendations on reddit might work.
this is definitely worth testing. although may be a temp loophole
 
Can experiment with it for sure. But then again, nothing can be said with certainty. I would use it to grow personal sites first than to directly use it for sites of the clients.
 
With the AI, honestly, we're now better off to be addressing AEO, since now the AI is the new Google. By "the new Google", what I mean is, back when Google was new, you might remember that many people just typed in questions into the search bar. While they surely still got answers, it took a long time before people returned to typing in a few keywords.

The AI, and pick whichever you like as an example, the average user is using exactly the same way; Asking it questions. Given some of the problems with AI the average user isn't aware of, how to avoid those problems, or how to tell when they are occurring, I doubt Google's going to become a thing of the past in the next decade or two--but I think looking at how those new to the internet used Google in it's youth would be a good way to estimate how user behavior will adapt to the AI itself.

Still, I'm a fan of combining tools, so I humor the idea, but I'd rather do some research myself. I've used the AI to generate what I like to call the "scaffolding" to copy for clients who work in extremely niche fields, as it can save what may have been significant amounts of time in research to write the bare basics. The upside is, while I work off it's generation, I do learn about the niche, my client is able to inform me if all the information is correct, and that allows me to really get into the SEO in terms of the target market. For pure giggles, if I had a client who wanted this help and their company makes perfume, well. . .I can market and advertise that, but let's pretend this company makes custom perfume. Sure, I can market a perfume already made, but marketing the idea this company will make a perfume for you is a lot different. Someone very dedicated might want to know about the types of chemicals in perfume, and I never took chemistry; Once I ask the AI to write about it, though, I'm going to learn, provided that's the topic. After that, for me, it'd be a lot of manual research into SEO terms, trends, etc. related to that.

Granted, I'm mainly approached for creating content with SEO (Blog posts, pages, meta), and I apply a huge amount of psychology to my approaches; In the above example, I'd assume my target audience are females, have high-school to advanced degrees, and are either already employed/stay at home for family/likely do not want or are able to make said perfumes themselves. Female consumers are intense in terms of research, though, so odds are they'd be looking to make sure chemicals used aren't harmful, aren't tested on animals. This combination is a huge market, and an extremely niche one. This isn't to say I'd write only for such consumers, but knowing all this, I'd be cross referencing with sources like the EPA's list or products known to contain harmful chemicals, what those chemicals are, how common it is to run into them. . .and why XYZ's Custom Perfume Creations is the safest choice, because XYZ doesn't ever use those chemicals.

I'm excited about AI but, it doesn't do the thinking for you, and that's the real problem and limitation I see going forward that needs to be kept in mind. In the end, it's still using things like Google, even if the user is being sold the feeling they aren't using something like Google. Since feelings are so often what we're really selling, again, to me, this goes back to consumer behavior and research, but I won't pretend I have any concrete answers. All I can say is this:

It sure helps with writing copy with SEO for some of the most odd, mind-numbing topics a lot faster, and more accurate, than without it. . .but if you're letting it do all the writing, you're going to sink, not swim.
 
Back