Main Body
278 Are Search Engines Penalizing AI-Created Content?
Up until the early 2000s, tricking Google by using tactics like keyword stuffing may have been commonplace, but the search engine runs a pretty tight ship now. In the recent ‘Reddy Book Club’, Google has reiterated what most good SEOs have known all along: create valuable content with the user at the center. The ‘trick’ to ranking in SERPs is not about outsmarting Google, but about working with it to add genuine value and provide real answers to users.
And yet, today the internet is abuzz with just one question: can Google filter out AI content? And will you be penalized for creating AI-generated content? The answer is yes and no.
Why Yes?
Post the helpful content update, the truth is that some AI content has definitely been hit. In fact, Neil Patel did an experiment where he tested 2 websites after the ‘Helpful content’ update. The one that only had AI-written content saw traffic drop more than 63% than the control website where humans edited the generated content.
Reading the first blog, you can understand that even if you didn’t detect that the text was machine-written, it offers very little value to the reader. It’s a mish-mash of sentences with very low originality and almost no excitement factors.
John Mueller, Google’s Search Advocate, made no claims of Google being able to detect AI-written content but did say that completely AI-written content would come under the category of automatically generated content which is against the webmaster guidelines.
“For us, these would, essentially, still fall into the category of automatically generated content which is something we’ve had in the Webmaster Guidelines since almost the beginning.
And people have been automatically generating content in lots of different ways. And for us, if you’re using machine learning tools to generate your content, it’s essentially the same as if you’re just shuffling words around, looking up synonyms, or doing the translation tricks that people used to do. Those kind of things.
My suspicion is maybe the quality of content is a little bit better than the really old-school tools, but for us, it’s still automatically generated content, and that means it’s still against the Webmaster Guidelines. So we would consider that to be spam.”
Why No?
This begs the question: How accurately can Google detect AI-generated content? Mueller makes no such claim.
However, content that you generate using AI tools, without any intervention has a certain pattern of words that is easy to detect. Read “How does AI writing work?” to learn more about why this happens.
Short-text items are generally very human-like. But when combined, without human help will read odd to humans and usually offer no real information making the content very low-value.
This essentially means that AI-generated content that is thin and spammy, could very well be on Google’s radar. But if you’re being intelligent about using this powerful technology to create content that stands out – you will be okay.
How to do AI content well
Are you or your writers treating AI content as an arch-nemesis? Do you believe that AI will be the reason you may not have a job tomorrow? If yes, then you’re probably going to deprive yourself of some of its amazing capabilities – capabilities to help you do your job better. Or at least make parts of your job easier to do.
It’s not Human vs AI but Human with AI
AI writing has become crazy with the barrage of tools offering to write your blog post for you. Analyzing Peppertype.ai’s 800mn generations across 65+ content types, we found that it’s great at
- Being an idea-brainstorming partner
- Expounding or simplifying sections of text
- Providing variations of content within seconds
- Short form text
A human-in-the-loop approach
We’ve now established that AI isn’t quite ready to take the driver’s seat, but it can be a superpower if used correctly. The quality of content created is great and very readable. In fact, Peppertype.ai data shows us that with the right nudges, AI+human content creation can decrease as much as 60% of the time spent writing content.
The approach that works well with AI writing that engages readers, doesn’t attract negative penalties on Google, and helps positively in SEO is a human-in-the-loop approach.
To summarise this well through an image from stanford.edu, the approach is meant to be a constant feedback system to the AI making the net output a lot better, while making the AI smarter over time.
