With the surge in AI writers like SEO.ai, Jasper and Copy.ai that all 10x the output speed for most human writers and marketers, the discussion around Google's stand on content produced by AI has increased.
Especially since an interview with John Mueller from Google in April 2022, where he compares artificial intelligence content with that automatically generated content.
Something which at the time was clearly against Google's Google Webmaster Guidelines.
This interview with John created a massive surge in articles describing how Google was against AI content and that the search giant potentially would give penalties to content consisting of these types of text and content.
Examples of this debate were:
- Google doesn't want your AI-generated SEO spam content
- A fierce battle is raging between Google's search algorithm and websites that use AI to churn out content
- Google Declares That AI-Generated Content Violates Policies
What's the argument for Google giving penalties to AI content?
Google has a history of penalizing websites with low-quality or auto-generated content.
And with good reason.
Sneaky SEO professionals have always tried to game Google's ranking algorithm long as SEO has been a discipline.
I can provide a personal anecdote to illustrate this point.
In my early days in SEO, around 20 years ago, one of the Black Hat strategies was programs that could scrape other pages with text and then copy-paste small sentences in an almost endless loop in order. This created new text pages with the original bits scrambled, so it was not just directly duplicate content.
The output of these content spinners was terrible gibberish - due to the limited quality of what could be coded back then - and the text had no actual meaning for a human being.
But Google's algorithm at that time was equally bad, and this type of content could rank in the SERPs for a limited time on keywords with low competition just because the text contained the right keywords.
So the battle between SEO profs and the Google engineering team has been a constant battle, with Google launching more and more measures and updates against these tactics to ensure users were given great search results.
The algorithms have evolved
As AI content so far has created output that falls in both of these initially mentioned categories - low-quality or auto-generated content - it makes sense that Google potentially should also penalize AI-generated content today.
But as you may already have experienced either here at the SEO.ai platform or using a tool like ChatGPT, the quality is now high, and it can massively improve the quality and how helpful content a writer can create.
Either by facilitating better content with relevant headlines and content suggestions - or simply by freeing up time for more research as some of the writing tasks are automated and speed up.
It might be difficult to fully grasp the techical achievements behind the different models, but by looking at just the different sets of training data they are trained on, one can understand the exponetial improvement:
- GPT-1 was trained on 7000 unpublished books (and the model had 117M parameters)
- GPT-2 was trained on 40GB of text data from over 8 million documents (and the model had 1.5 billion parameters - approx 10x the predecessor)
- GPT-3 was trained on 45 TB text data from multiple sources which include Wikipedia and books (and the model had 175 billion parameters - approx 100x the predecessor)
Google's position is now changed - AI is no longer to be avoided
As mentioned initially in this post, Google has been against automatically generated content which one could argue AI text falls under.
See the screenshot from April 7th, 2022 of the old Google Webmaster Guidelines at developers.google.com/search/docs/advanced/guidelines/webmaster-guidelines (no longer active link - can be accessed with the Wayback Machine)
Already at the end of April 2022, Google updated this formulation (maybe after the heavy coverage in the SEO community of the John Mueller quotes) of how to avoid the techniques with automatically generated content to be followed by an "intended to manipulate search rankings".
Since then Google Webmaster Guidelines have been revamped and become Google Search Essentials, and now there is no clear guidance in the Google guidelines on avoidance of solely automatically generated content.
Now it's explicit that it's just the "Spammy automatically-generated content" that should be avoided.
Of course, this can also cover AI written content if one uses these kinds of tools to generate content without having the user in mind but only SEO purposes:
In the Spam Policies section Google is most clear on what to avoid (and thereby what also to keep in mind when when writing with AI tools):
- Text generated through automated processes without regard for quality or user experience
Danny Sullivan, Public Liaison for Search at Google, also made it clear back in November 2022 that it's not against Google guidelines to use AI to create content, as long as the output is written for people and not just the search engines.
UPDATE January 2023:
John Mueller, who had previously been skeptical about how to view AI-generated content, has now become more accepting of its use. He recently confirmed in a Mastodon thread that AI can be used for titles and descriptions.
But even without this change, Google might have a practical issue as well. Can Google detect if text or content is generated by AI?
The short answer is that they might have been able to do this, but as new technologies emerge which is yet more and more advanced and thereby more and more human-like it seems unlikely in the future.
How to create AI content Googles love
In short, Google doesn't care if the content is created BY people, as long as it's created FOR people.
How do you then achieve this?
Well, Google has given quite good instruction in their 2022 helpful content update where the headline also is "Focus on people-first content".
Here's the list the following:
- Do you have an existing or intended audience for your business or site that would find the content useful if they came directly to you?
- Does your content clearly demonstrate first-hand expertise and a depth of knowledge (for example, expertise that comes from having actually used a product or service, or visiting a place)?
- Does your site have a primary purpose or focus?
- After reading your content, will someone leave feeling they've learned enough about a topic to help achieve their goal?
- Will someone reading your content leave feeling like they've had a satisfying experience?
- Are you keeping in mind our guidance for core updates and for product reviews?
Follow these, and you should always be home safe, AI writing tools or not.
Conclusion: What is Google's current stance on AI text?
So what is the conclusion and sum up on this?
The debate surrounding Google's stance on AI-generated content has increased with the rise of AI writing tools.
Google has historically penalized websites with low-quality or auto-generated content.
Google's position on AI-generated content has evolved, and it now specifically targets "spammy automatically-generated content."
.svg)
