Does A.I. content get penalised by Google?

It may come as a huge shock, but the answer to this one is a whole lot of "it depends". It all comes down to how you use AI content, how useful it is to your readers and whether you're using it as an aide or replacement for human experts.

(TL;DR: Probably not unless you're spamming your site on a large scale. But that doesn't mean that your AI content is good enough to rank)

First of all, let's stop saying "penalise"

One of my pet hates in the SEO industry is the misuse of the word "penalise". I see it being used on am almost daily basis by those who evidently don't really know what it means. A Google penalty (whether algorithmic or manual) is when Google sees something on a website that violates its terms and conditions and takes action to stop it appearing in search results.

Most of the time, if a web page struggles to rank or get indexed its not because of a penalty. It's because the content is shit. This is a very different issue and should be treated as such.

This differentiation is particularly important when talking about AI content. There are cases where AI-generated sites provide no value and are treated as spam. There are other cases where an AI-generated article just isn't very good and needs improvement.

How does AI content work?

AI content generation tools use frameworks such as GPT-3, which takes existing content then applies machine learning it rewrite it to your specifications. Think of them as sophisticated search engines which can find all the information your request from the internet, then stitch it all together into a readable piece of text.

Since Google's goal is to give you the very best content to fufil the intent of your search, you can see where the issue is here. Why would Google rank your content above others in search results if it offers nothing new? This is demonstrated in the following quote from John Mueller with regards to indexing:

What ultimately works best is that you prove to Google (and users) that the updates you’re providing are valuable: unique, compelling, high-quality, and not something that’s already published elsewhere.

So if your "well-researched" article is just piecing together work that someone else, should Google care enough about it to index it? Well, maybe. We'll come to that in a bit.

To learn more about how AI generated content works, click here for a free trial with jasper.ai.

Pros and Cons

AI generated content can be a very powerful tool when it comes to creating content quickly and building out websites, but it isn't perfect. Below is a list of pros and cons of using AI content for your website.

 Pros Cons

✔️ Create content quickly

✔️ Scale up websites quickly

✔️ Cheap

✔️ Anyone can use

❌ Can't create new ideas

❌ Struggles to create longer articles

❌ Lacks individual writing style

❌ Can make creators lazy

What do Google say?

Google's stance on this has softened as of late. As recently as April 2022, John Mueller said the following about AI generated content:

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 for us it’s still against the Webmaster Guidelines. So we would consider that to be spam.”

This sounds pretty damning and seem like a nail in the coffin for AI content in the eyes of many SEOs. However, they seem to have softened their stance in recent months on the subject.

In August 2022, Google announced the Helpful Content Audit. Many SEOs thought that this would primarily target AI content, signalling the death knell to this form of content. Yet, when you read the details of the update this never happened. In response to an article from Mike King about AI content, Google's Danny Sullivan said the following:

So, reading between the lines here we can infer that Google doesn't mind if you use AI content, but it does mind how you use it.

What do the experts say?

As the old saying goes; opinions are like noses; everyone has one (this isn't the actual proverb, but I can't say what it actually is as Google apparently dislikes profanities). Many SEOs have opinions about the value of AI generated content, but I'd prefer to focus on what the data says. However, the data is, well, mixed.

 

One of the best case studies I've seen is from (hear me out) Neil Patel. He monitored rankings from 100 AI generated informational sites. Half were human-edited and half were copy and pasted. After a Google spam update, most of the sites lost traffic but the non-human edited sites lost considerably more. So this would imply that Google doesn't like AI content right?

Another case study I saw was from Nick Wilsdon who created 4,000 new categories on an ecommerce site and use AI to create copy on each of these pages. The result? The site is absolutely flying.

So why are these examples so different? Its comes down to how the sites are using AI. In the latter example, the copy needs to describe the contents of the page and use persuasive language to convince users to engage with the site. And lets face it, does anyone actually read ecommerce category descriptions? In the first example, the sites are mainly informational pages. These sites need to be well researched and trustworthy, which is difficult for AI content generators to do.

Another issue with AI is content length. The longer the copy, the lower the quality will be. In Neil Patel's example, most of the articles are only 500 words as AI tools can't really produce longer articles at scale. 500 words these days is nothing and is unlikely to have the amount ot detail necessary to compete with more trusted sites in Google search results. However, category copy is usually a lot shorter so is much easier for AI tools to get right.

What are the implications of AI content?

As you can imagine, SEOs have a huge influence in what appears in Google's search results. I might get crucified by my peers for saying this, but this is not always a good thing. You see, the primary goal for most SEOs is to get the traffic which makes money. It's not really the job of most of us to make the internet a better place. Many SEOs rely so much on optimisation tools like Surfer or frase that they don't even realise that there is a person searching for their content who wants to know the answers to their question and to trust its written by someone who knows what they're talking about.

Meme showing issues with quality of Google's search results

In recent years, I've seen a few people recommending to use "site:reddit.com" when searching for questions in Google. This is because Reddit is full of genuine experts who love to share their opinion. Whilst there are many factors that feed into Google Trends, the demand for reddit over time reflects this.

Google Trends interest for Reddit over time
Whilst this issue is not exclusive to AI content, it does raise a serious concern for the industry. At the end of the day, SEOs only have a job whilst their audience trusts the results. Whether its AI content, or outsourcing to generic writers for 4¢ a word, many SEOs use tactics that do not result in expert-written content ranking in Google. The long term consequence is that Google may lose their dominance in the search industry and users may look for other solutions to get the answers they need.

So what does this all mean?

 In summary, whether or not Google devalues AI content comes down to what you're using it for. Let's break it down to different ways you can use AI generated copy.

  • Short snippets of text, such as generating meta descriptions. This is a good example of where AI can add real value to a site - Risk: Low
  • Short paragraphs, such as category descriptions. AI can work really well for these as long as provide a solid brief - Risk: Medium
  • Paragraphs within blog articles. Again, getting the brief right is the key here, AI can be really effective if you instruct it properly. Just be cautious that content requiring a high standard of research may not be be of a required standard - Risk: Low/Medium
  • Creating the framework of an article to be human-edited. This can be a real time-saver and can help those who are experts, but not necessarily good writers. However, this tactic can make it tempting to cut corners - Risk: Low/Medium
  • Creating entire articles to be human edited. This comes down to how well you scrutinise the articles, putting too much faith in AI content generators can potentially compromise the quality of the output - Risk: Medium/High
  • Creating articles at scale with minimal editing. This is a common tactic which is working for some creators, but is extremely high risk. It's a tactic that can work well if your intention is to manage a churn and burn site, but in the long term your site will probably suffer - Risk: High
  • Creating YMYL content at scale with minimal editing. An example of this is providing medical advice without putting the effort in to ensure that this advice is correct and trustworthy. Good luck with that one, Google is likely to come down on you like a ton of bricks - Risk: Very High

In my opinion, there is a ton of human-created content out there which is garbage and offers no benefit over AI content. If you're using generic freelance copywriters on 4¢ a word and think that the only SEO that is required is using Surfer, then the content will not be great quality. If that is your tactic, then you may as well use AI.

However, for well-written and thoroughly researched content which is likely to stand the test of time with future algorithm updates, AI should be seen as a writing aid and not an outright replacement for human writers.

The best way to learn is to have a go. Click here for a free trial with jasper.ai.

(We use affiliate links in this article to help pay our bills, but the analysis is completely impartial)