AI Generated Content: Google’s Stance, Evidence, Theory & Ideas

 Ever since AI Generated content became easily accessible, people have been wondering whether the easily acquired content can help websites in ranking. The short answer is, Yes and No. The question you ask isn’t simple, but you should have your answer by the end of this article, along with an idea of how to proceed further with content generation from now on.

Evidence: Google’s Only Concerned with Quality

Google has been quite adamant about things when it comes to content and makes sure that anyone and everyone in SEO is aware of the importance of EEAT. Altho0ugh Google states that it looks for helpfulness in content, the evidence says otherwise.  

Many believe that AI-Generated Content was responsible for sites that Google deindexed during March 2024 (back when the update was rolling out). Multiple Reddit threads like this one popped up with people complaining about their sites getting deindexed. The chatter online that this update might be specifically targeting AI content was buzzing, as many websites got a manual action penalty along with being deindexed for around 2 days. Later on, some of these sites reported that they had gotten indexed again with the manual action penalty being removed. 

However, when John Mueller’s website got deindexed, everyone knew that it wasn’t AI content that the update was specifically targeting. But speculations aside, the most common factor with websites Google hit with manual action and deindexation was the label of “Pure Spam”.

Pure Spam, according to Google is defined as follows:

The main focus that Google had behind was tackling spammy, low-quality content. So, it was never about the AI, but rather the quality. During this time, Natzir pointed out the change in Google’s Quality Rater Guidelines on exactly how Google will be determining content quality. 

Theory: Why AI Content Websites Were Affected?

With Google cracking down specifically on lowest-quality content, the classification for low-quality content has obviously become a lot more strict than it was previously. Currently, Google can’t fact-check every blog its crawlers come across. Moreover, the frequency of blogs published every day has only increased with the advent of ChatGPT and other AI tools.

You can get AI content in mere minutes. So the larger task isn’t publishing but rather editing and fact-checking. And it is one of the major reasons why websites using AI-generated content put up disclaimer clause in their policies. The clause stated about the possibility of inaccuracies due to use of AI in content generation. However, this put Google at a disadvantage in discerning whether content pieces are fact-checked and reliable or not.

Thus, more websites with AI content were seemingly affected compared to human-written websites. Here’s a picture of the recent addition to Search Quality Rater Guidelines to drive our point home.

This does not mean that Google is against AI content, but it does showcase that Google values quality over quantity. Google already clarified its stance on AI-generated content a year ago, stating that it doesn’t matter who makes the content as long as it is helpful.

AI vs. Human Content: The Core Competence

Of course, you will find pros and cons with both humans and AI in content generation.

If you choose to go with AI content, then you’ll be getting:

• Lowered Costs: Since AI is practically free, content generation becomes cheap. Employing people to edit AI content should still cost considerably less than having human writers.
  
• Time-Benefit: AI can generate content in a matter of seconds. Even if you’re going for a large piece, AI won’t take more than a few minutes to give you a complete piece.
  
• Scalability: Thanks to AI, instead of having to spend weeks creating a limited number of blogs, you can virtually create a hundred blogs in a single week without any hassle. 

Now let’s discuss how Human writers might be a better option:

• Authenticity: Humans may cost more and take more time, but their content will likely be more authentic. With proper research and references, you know that each piece of content created is true, in every sense of the word.
• Creativity: One of the few things AI still falls short on is creativity. Sure, you can get run-of-the-mill stuff with AI easily, but writers are more in tune with your audience and can use terms, phrases, and even metaphors that perfectly align with your target audience. Even with a large amount of processing, right now, AI still can’t come close to the creative ideas humans generate. 
• Quality: When it comes to quality, humans still have the upper hand. Human writers can establish a rapport with the target audience, whereas AI’s tone will mostly be generic until edited. This also leads to better trust in your content from your target audience.

What Does Research Say?

Well, for now, the results seem divided from study to study on whether AI content is better or whether Humans still rule the realm. An MIT study indicates that readers have a bias toward human content, often preferring content after finding out it was written by humans instead of AI. However, AI content performing slightly better at delivering satisfaction to 1,212 participants does deserve acknowledgment.

AI-augmented content came next. Here AI crafted the content draft, with humans editing it. AI triumphed even when it came to copywriting and creating perceived value. The AI generated content came via ChatGPT-4, the premium version.

Do keep in mind this study is talking about satisfaction and not rankings. When it comes to ranking, we looked up another case study by Marketing Insider Group. This study was using the relatively inferior ChatGPT 3.5 to generate content. The study used AI, the Hybrid Model (AI content augmented via Humans), and Humans to write up 60 Blog Posts (20 each).

Marketing Insider found out that Human content surpassed AI significantly, in most areas. Also, since the study focused on rankings, it shows that AI content does rank. One of the main findings of the study was that AI lacked the depth that Human-written content could provide. However, it also pointed out that the Hybrid Model showed potential. 

They concluded that AI content struggled to keep consistent rankings. The study indicated that AI content often fell down the SERPs instead of climbing up. While the Hybrid content model did have its moments of success, in some instances even surpassing human content, these moments were quite inconsistent. 

On top of that, when it came to Web Traffic, Human content easily outperformed AI and Hybrid content. AI did find success in one post that brought a lot of traffic. However, a 5% success rate isn’t something many would find interesting. 

However, AI did get the benefit of visibility thanks to its 1 excellent post, resulting in high visibility initially. But this advantage declined as the study progressed. The Hybrid content model occasionally did surpass human content but again these achievements were inconsistent.

Ideas: What’s the Best Way Right Now?

What performs better seems to vary from study to study. So, the best strategy is to have more quality checks in place. While AI will provide you with scalability, it may come with risks and strings attached.

Fact-check everything in your content if its made with AI. Also make sure your content pieces are as helpful as possible so they rank better on the SERPs.

With human written content, your writers’ expertise will decide the quality. However, you will have limited quantity. In case you pick the Hybrid model, then you’ll get the benefit of scalability without compromising quality.

One thing we do recommend is experimentation. If you work with human writers, testing AI should provide you with comparative results. But if you are getting rankings and results via AI, you don’t need changes.

To conclusively emphasize, our point is, “If it ain’t broke, no need to fix it. And if it is Pivot!