8 AI Detection Tools Recommended by Ahrefs

As a Content Editor or SEO Specialist, there are times when I feel a little concerned about hiring freelance writers outside of my regular team.

One question that sometimes comes to mind is:

“Was this article actually written by the writer, or was AI heavily involved in creating it?”

With AI becoming increasingly accessible, this has become a common concern for many people working in content and SEO.

Fortunately, there are now plenty of AI detection tools designed to estimate whether AI was involved in creating a piece of content.

These tools can be useful for filtering and reviewing articles before publishing them on a website.

I spent some time looking for reliable AI detectors and eventually came across an article by Ryan Law, Director of Content Marketing at Ahrefs. In his article, he tested several popular AI detection tools and shared his findings.

Based on his recommendations and my own experience trying some of these tools, here are several AI detectors worth checking out.

Recommended AI Detection Tools

Let’s start with the first one.

1. Ahrefs AI Detector

If you want to identify not only whether AI may have been used but also potentially which AI model was involved, Ahrefs AI Detector is one of the more interesting options.

According to Ahrefs, its AI detection technology can identify patterns associated with specific AI models, including models such as GPT-4o and Meta’s Llama.

The tool is built on Ahrefs’ proprietary technology and trained using a large amount of web content.

What makes Ahrefs AI Detector particularly interesting is its connection with other data available inside the Ahrefs ecosystem.

Instead of simply checking whether a piece of content appears to be AI-generated, you can potentially use AI detection data alongside SEO metrics.

For example, you may be able to explore things such as:

• Which AI models appear to perform better when creating content.
• How frequently competitors publish content that appears to use AI.
• Which AI models competitors may be using.
• How much AI-generated content appears across search results.
• Whether there is any relationship between organic traffic performance and AI usage.

Another useful aspect is that Ahrefs provides access to other page-level data, such as historical content changes, backlink information, and estimated organic traffic.

The downside is fairly obvious, though.

You generally need an Ahrefs subscription to access the tool and its wider set of features.

2. Copyleaks

Copyleaks is another option worth considering if you want to check whether AI may have been involved in creating a piece of content.

Ryan Law described the tool as solid and effective when it comes to AI detection.

From my perspective, the interface also looks relatively beginner-friendly.

The process is straightforward. You simply open the Copyleaks website, paste the content you want to analyze, and run the scan.

I think this can be useful for Content Editors who want to perform an additional review before editing or publishing an article.

Another thing I like is that Copyleaks offers integrations such as a Chrome extension and Google Docs add-on.

That means you may not always need to visit the main website every time you want to analyze a piece of content.

3. GPTZero

GPTZero offers a slightly different experience because it provides more options for exploring and analyzing content.

One convenient feature is the ability to upload or work with documents without always needing to manually copy and paste everything.

Similar to Copyleaks, GPTZero also offers integrations for Google Docs and Chrome.

The process is relatively simple. You can select the content you want to check and use the available AI scanning feature.

There is a free version available, although some features are limited.

If you need more access or advanced functionality, there are also paid plans available.

4. Originality.ai

According to Ryan Law’s testing, Originality.ai also delivered relatively strong results when detecting AI-generated content.

In many cases, the tool was able to identify content that appeared to have been generated by AI.

However, Ryan also pointed out one potential weakness.

The tool may struggle when AI-generated content has been heavily edited or rewritten by a human.

In those situations, the detector may classify the content as human-written rather than AI-generated.

This is one of the reasons why I personally don’t think AI detection tools should be treated as absolute proof.

A detector can provide useful signals, but it may not always tell the full story behind how an article was created.

5. Scribbr

Scribbr can also be used as an alternative for detecting possible AI involvement in content.

However, based on Ryan Law’s observations, the tool appeared to be more cautious when analyzing more complex content.

Instead of confidently assigning a high AI score, it may be more conservative with its results.

On the positive side, Scribbr is relatively simple and easy to use.

The interface doesn’t feel overly complicated, making it accessible for people who simply want to paste content and run a quick check.

At the time of writing, I also couldn’t find an official Google Docs add-on for Scribbr, which may be something to consider if you prefer working directly inside Google Docs.

6. ZeroGPT

ZeroGPT is another popular name in the AI detection space.

However, Ryan Law found that its results were not always consistent.

According to his observations, ZeroGPT sometimes struggled to properly classify content with low or partial levels of AI involvement.

The results could also be inconsistent when comparing completely human-written content with hybrid content created using both AI and human editing.

I also tried testing one of my own pieces of writing that I considered completely human-written.

Interestingly, the tool still detected around 20% AI involvement.

This is a good example of why AI detection scores should probably be treated with caution.

A percentage score doesn’t necessarily mean that a specific percentage of the article was literally written by AI.

7. Grammarly

Most people working with English content have probably heard of Grammarly.

It’s widely known as a grammar and writing assistant, especially for English writing.

However, Grammarly also provides features related to AI detection.

According to Ryan Law, Grammarly was not among the strongest performers when it came to accurately identifying AI-generated content.

It reportedly had difficulty recognizing certain AI patterns and distinguishing hybrid content created through a combination of AI and human writing.

Interestingly, when I tested one of my own articles using Grammarly, it returned a result of 0% AI.

Of course, this was only my personal test and shouldn’t be considered a scientific benchmark.

Still, it was interesting to compare the results with other AI detection tools.

8. Writer

Writer received one of the weaker evaluations in Ryan Law’s testing.

According to his findings, the tool frequently misidentified content generated by AI.

The AI detection signals appeared to be relatively weak, particularly when dealing with content that was supposedly generated entirely by AI.

That said, there is nothing wrong with trying it yourself and comparing the results with other tools.

Once again, I tested one of my own pieces of writing that was written entirely by me.

The result classified it as 100% human-generated content.

This shows just how differently AI detection tools can evaluate the same type of content.

Choose AI Detection Tools Wisely

The rapid growth of AI can be seen from two different perspectives.

On one hand, AI can make content creation faster and more efficient.

On the other hand, AI can also be used as a tool to improve research, structure ideas, and potentially create more useful content for readers.

For that reason, I don’t think the purpose of AI detection should always be about completely banning AI.

Instead, these tools can be used as an additional layer of quality control.

The list above is largely based on Ryan Law’s evaluation published by Ahrefs.

In his testing, he reportedly analyzed nine different articles consisting of:

• Three articles written entirely by AI.
• Three articles written entirely by humans.
• Three hybrid articles created with approximately 50% AI and 50% human involvement.

From my side, I also tried testing several of these tools using my own writing.

However, my personal tests were much more limited and shouldn’t be compared to a structured benchmark.

If you’re planning to use AI detection tools, I would recommend testing them yourself first.

Try using different types of content, including:

• Content written entirely by you.
• AI-generated content.
• Content created using AI and then heavily edited by a human.

This can give you a better idea of how each detector behaves.

Final Thoughts

AI detection tools can be useful, especially for Content Editors, SEO teams, and website owners who want an additional way to review content before publishing it.

However, I personally wouldn’t rely on a single tool or score as absolute proof that an article was written by AI or a human.

Different tools can produce very different results.

A piece of content might be classified as AI-generated by one tool and completely human-written by another.

That’s why I think it’s important to use AI detectors wisely.

Treat them as an additional signal rather than the final decision-maker.

At the end of the day, content quality, accuracy, originality, usefulness, and the value it provides to readers are still far more important than simply chasing a perfect AI detection score.

Reference:

Ahrefs, The Best AI Detectors (Tested)
https://ahrefs.com/blog/best-ai-detectors-tested