I’m sitting here on the Friday after Google released its latest update aimed at tackling spammy low-quality content on Search.
What I am seeing is a lot of people declaring that Google has finally gone after AI generated content and this is the end of generative AI.
With everything I am about to say, keep in mind that we are only a few days of what Google has announced may be a 4 week rollout.
Has Google gone to war with AI generated content?
In short, yes, but only bad AI content.
I’m waiting until next week for Search Console data to update, but I will include some screenshots here of cases where clients are using AI generated content and that content is doing just fine.
Now to be clear, on these projects we are using AI in smart, targeted ways. For example, a few clients have built out knowledge bases on their sites, something I talked about in a previous note.
One client has about 3000 templates they offer. It didn’t make much sense to pay writers to create content for those pages including descriptions, use cases, etc. We did it with GPT-4 and some excellent prompting.
All of these sections are doing just fine, despite being 100% AI generated.
So what is all the buzz about? I saw that (Insert SEO YouTuber Here) said their sites got hit
And this is largely what has happened so far.
There have been quite a few YouTubers and other SEO influencers on social media platforms bragging about how they were spitting out thousands of pages of AI generated content and jumping to the top of the SERPs.
Many of us saw Jake Ward’s “SEO Heist” on LinkedIn and Twitter, which worked…
…right up until it didn’t.
(Sad thing about that one was it was client work rather than testing on his own site.)
[Note, this one happened before the recent update.]
But since the update many other YouTube SEOs have found themselves targeted in similar fashion.
Julian Goldie saw every site he shared publicly get a friendly little de-index from Google.
Some guy who chugs straight workout powder for some reason saw the same with all his sites.
There are many other SEOs who have been bragging about and sharing sites where they were sticking it to Google with AI content.
This time Google didn’t stop with the sites these SEOs had shared publicly. They also de-indexed any other site using GSC on the same Google account.
They made an example of them. Much like they did to Kyle Roof after his Plano rhinoplasty experiment.
That’s the thing about Google. When you publicly shame them, they will act.
This has been a running standard practice with them. Anyone remember the JCPenney incident all the way back in 2011?
JCPenney got penalized for using shady link practices, but it wasn’t because Google caught them.
The New York Times published an article in which they found that JCPenney was using unnatural links to boost their search visibility.
Immediately afterwards, Google removed JCPenney’s website from showing up in search results.
They were shamed into action.
So is it safe to use AI generated content?
Yes, but like anything else, it is best done in moderation and with a targeted plan.
If you start blasting out thousands of AI pages, you are significantly increasing the risk of something bad happening to your site.
But Mike, you said Google only targeted the SEOs on YouTube and other social channels… I don’t have a YouTube channel and I don’t share anything I do publicly. I would be safe to go full blast, right?
First, I didn’t say that they only targeted them. They have just been the most visible targets so far. That doesn’t mean there were not a lot of other sites hit too.
Fresherslive.com was one I saw people talking about on Reddit that I had never seen mentioned anywhere before.
There are certainly other sites that were hit.
Second, what Google does manually one day, they strive to do through their algorithms in the future.
I mentioned the story about JCPenney above. There had been other similar instances like that, but this was the biggest company made an example of.
Google started making a lot more noise about shady link practices. A few months later that year we saw Google sending out notices through Google Search Console (Webmaster Tools at the time) for unnatural links and additional notices for sites that were potentially buying or selling links.
Then just over a year after the JCPenney story broke, Google released Penguin.
I see a lot of similar patterns right now. Google is certainly making noise about people who are using high quantities of low quality content. It just so happens that the easiest way to produce low quality content right now at scale is through AI.
We are likely to see something similar to Penguin in the future to address this.
I do not believe they are going to target all AI content, just like Penguin didn’t target all links. However, if you remember with Penguin there was collateral damage.
So you are saying to stop using AI content then?
Not at all. I’m just saying to be careful and make sure you are using it as a writing tool to produce accurate, quality content. Make sure the content is edited by an actual person.
I think using personas when creating AI content is a big differentiator between generic obvious AI content and creating something more engaging and on point for your brand.
I’ve already shared numerous tips here for how to do that including one just a few weeks ago about helping AI to write better for your brand.
I will of course continue to share AI tips as I am constantly refining my own practices.
If you are using generative AI content, I would recommend taking a hard and honest look at what you have produced. Try to remove your personal bias as much as you can. Even have friends or colleagues read your content.
Is it really something useful to readers and searchers or is it the exact same answer I would get out of ChatGPT by just asking it a simple question?
If it is the latter, you may want to think of ways to improve it.
