Anyone using AI content detector?

Guestwriting

Elite Member
Jr. VIP
Joined
Jul 30, 2012
Messages
7,641
Reaction score
1,476
Hi,

Anyone using this tool: https://copyleaks.com/ai-content-detector Does it give a real result?

Yesterday, one of my clients told me that our content was AI-detected. I also checked on the same day, and it showed me the same error.

Today! We wrote another piece of content, and it again shows the same error: "AI content detected." When I checked my old content again, it said "This is human text". I was caught between the client and the writer.

Valuable input about this tool is welcome.
 
I'm not sure if it inaccurate but I usually create content to bypass copleaks.

So, it can be fooled. I've actually never put in actually human written material in.
 
The situation is really ambiguous, and it's hard to understand how it actually happens.
 
These tools are not 100% accurate. It is still lacking a lot.
 
It's not cheapo. I prefer the credit model-based platforms. I've tested some texts and it doesn't seem too bad though. It's like Originality, but Originality does have a credit-based system.

I've also put a AI generated content and it tagged as full Human text so we must understand this is kind of tricky. But I currently believe that is better using a tool that not using any tool at all. At least you could get a glympse of what a machine could be recognizing to some extent. Better to have glasses that to go almost blind in this aspect.
 
I just ran several articles through copy leaks that were 100% Ai and each one came back as human
 
I use them, to make sure my GPT generated content comes up as human written
 
None of those tools is reliable, even 30 % is a stretch. Pure guessing.
 
Those AI detector is a kind of scam, none of them show how it works. It's very hard to make a tool like chat GPT, and to make a tool that can detector AI writing in chat GPT, well, that tool is basiclly have to be better than the GPT itself. Everybody has their own way to write, no one can know if it's AI just base on writing style. Even some arts drew by AI is mistaken for human product.
 
I used it a few days ago, including paid detectors. Each detector flags the content as Ai in one way or another.

They even detect Ai content in articles written before Ai was even a topic, and the articles weren't spun or translated or anything else.
 
It's debatable same as the plagiarism detector.
Considering you are a service seller, I think you need to tell your writer it needs to pass this plagiarism and ai checker tool. as a requirement.
Since that is what matter to the customer.
 
I used to but I heard Google doesnt care if it's AI, as long as it's good helpful content.
 
I use a wide range of AI content detectors.

The one from the writer.com website is particularly good.

I use a wide range of these tools to make sure that my stuff is rated 100% human as much as possible.

With that said, people publishing content shouldn't just focus on a numeric approach to their content.

What I mean by that is just don't be content with just thinking that just because it passes a human detection tool, then you can rest.

No.

You also have to make sure that your stuff is easy to read, it's easy to scan, and it's actually useful.

This is the key.

It has to be usable because at the end of the day, you have to deliver value to the lives of people you expect to click on an ad and somehow, someway put dollars in your pocket.

If that doesn't happen, then you failed to do your job.
 
i'm not exactly sure about this model, but the detectors i have used have varied wildly in quality; sometimes they will detect correctly, sometimes they wont. if you even change a few words around your results might look different.
 
Those AI detector is a kind of scam, none of them show how it works. It's very hard to make a tool like chat GPT, and to make a tool that can detector AI writing in chat GPT, well, that tool is basiclly have to be better than the GPT itself. Everybody has their own way to write, no one can know if it's AI just base on writing style. Even some arts drew by AI is mistaken for human product.
Two words: reverse engineering.
 
I have generated tons of content using AI in the past and used numerous tools, including Originality.ai. Let me tell you one thing; your content needs to be readable and engaging to the readers. AI-Generated content is bland, very generic, and follows a writing pattern.

If you somehow make it interesting with AI, content detectors will not make any difference even if it gets detected. AI content still ranks on google, and everyone knows it.

Why are sites getting approved by Adsense even though the content has been written by AI? They are getting reviewed manually. Right?

You basically need to understand how these tools work and the best example would be to look at Open AI's playground tool. If you go there and generate something, increase the frequency penalty and other settings to max and you will bypass every detector but the content will be a disaster.

AI or no AI, what matters is "Plagiarism." If everyone is using the same date/information, it will eventually generate the same article and that would be a problem.
 
Back
Top