Over the past few months, I’ve been tracking several of these AI content detectors. My focus was primarily on their ability to identify AI-generated content rather than plagiarism. The “industry standards” are GPTZero and Originality.ai, but several other tools, such as ZeroGPT, Quillbot, and Undetectable.ai, have also entered the fray.

With surprisingly minimal effort, I’ve managed to bypass the detection algorithms of all these tools.

In this article, I’ll explain how I did so and why the process isn’t as complicated as you might think.

Conflict of Interest

Each of these detection tools claims to be the "best" at identifying AI-generated content, but they also offer services to "humanize" that same content to help it evade detection. This presents a clear conflict of interest and they sell solutions to the very problem they claim to solve.

Tons of content has been produced comparing these tools , analyzing their accuracy and ranking them based on different factors. However, I always believed this is quite futile. Even if a tool excels in detection for a time, it won’t last long. Each new update to a language model forces these detectors to play catch-up, making it a perpetual cat-and-mouse game. Their effectiveness fluctuates, with many users reporting false positives and negatives, highlighting their unreliability.

The Paradox of AI-Detecting-AI

Most of these detection tools are AI-based themselves, which creates a paradox. Using AI to detect AI-generated content can never be 100% accurate, especially as language models continue to improve. This makes it increasingly difficult to distinguish between human and machine-produced text.

To understand why these tools struggle, it's essential to look under the hood at how they operate. Once you grasp the mechanics, bypassing their algorithms becomes much simpler. And that’s precisely what I’ve achieved, with consistent success.

How AI Content Detectors Work

AI detectors, such as GPTZero, rely on several key methods to identify AI-generated text. Here’s a breakdown of how they function:

1. Perplexity Analysis

Perplexity measures how "surprising" or unpredictable a piece of text is to a language model. Text that aligns closely with predictable patterns might be flagged as AI-generated, while more nuanced and unpredictable language often indicates human authorship.

2. “Burstiness”

Burstiness refers to variations in sentence length and structure. Human writers naturally vary their sentence structures, creating a dynamic flow. AI-generated text can be more uniform, lacking this variation, and detectors use this pattern to identify possible AI origins.

3. Repetition and Redundancy

AI models may inadvertently repeat phrases or ideas, a sign that detectors can pick up on. By analyzing repetitive elements, detectors can identify AI-generated content that lacks genuine variation.

4. N-gram Analysis

N-grams are sequences of words (bigrams for two-word sequences, trigrams for three, etc.). Detectors analyze these to check how often certain phrases appear. Overly predictable n-grams often signal AI authorship, as models rely on common word pairings.

5. Style and Tone

AI-generated writing can sometimes feel too consistent or polished. Detectors look for stylistic markers that might suggest a lack of nuance or creativity, elements that typically characterize human writing.

6. Training Model Comparisons

Some advanced detectors compare the input text against a database of known AI-generated content. They can also run the text through similar models to see if they produce matching outputs, which would suggest AI authorship.

7. Contextual Understanding

Detectors analyze coherence and logical flow. AI-generated text can be semantically accurate yet lack deeper contextual understanding, creating subtle logical gaps. Detectors flag content that is technically correct but contextually off.

By combining these approaches, AI detectors like GPTZero and Originality.ai attempt to distinguish between human and AI-generated content. But understanding how these tools work makes it much easier to outsmart them.

AI Detectors, Defeated

Armed with this knowledge, I’ve found it surprisingly simple to bypass their algorithms. By using a considerably simple prompt (just 11 lines long!), I managed to produce content that bypasses their detection mechanisms. The result was AI-generated text that passes as 100% human-grade content according to these tools.

Here is an example of some ChatGPT generated text describing global warming. Originality.ai successfully detects it as 99% AI:

I then re-generated the text through my prompt and this it’s now 100% human content (!!):

I then ran the same test on GPTZero with the original text from ChatGPT, not surprisingly it was successful in detecting it as AI generated albeit with less confidence than Originality.ai, at just 76% probability:

I then gave it the re-generated text from my prompt and surprise, it’s suddenly 98% human:

If you’re curious to try it for yourself, I made it available for free on the GPT store.

Simply paste any AI-generated content, and the tool will modify it to appear more human. You can then verify the results by running the output through your preferred AI detection tool.

The quest to detect AI-generated content is like trying to hit a moving target. Every advance in language models pushes detection tools to adapt, creating a never-ending game of cat and mouse.

PS: If you’d like to see the prompt, feel free to subscribe to my sub and you’ll get a link which contains it, as well as all the other prompts from my previous posts!

I’m building a tool that does exactly this. Basically once you set an initial prompt—like “daily inspirational quotes” or “fun facts about space”—the AI fetches new results based on that same prompt every day and delivers them straight to your inbox. No need to return and enter fresh commands daily—the AI continuously runs your initial prompt through your favorite platform (ChatGPT, Claude, Gemini, etc) to provide you with new content each day.

A simple use case to illustrate how it works: getting a daily dose of inspiration. You might set a prompt like “daily inspirational quotes” and then simply enjoy the results that show up in your email each morning. Every day, the AI will generate fresh motivational content from its language model, offering a new set of words to inspire you.

But this tool can do much more than deliver inspiration. Let’s say you’re a fan of a specific author. For example, you could enter “daily quotes from Maya Angelou,” and the AI will generate unique responses from its knowledge of her work. Whether you're drawn to a certain philosopher, writer, or thinker, the AI can tap into its vast database to create new quotes or sayings that fit your original request, keeping your favorite author in your daily routine.

The tool also shines when it comes to learning new things. If you’re passionate about a subject—whether it’s “the history of ancient civilizations” or “facts about climate change”—you can set a prompt for “daily interesting facts about space exploration” or “daily historical events.” Each day, the AI will generate and send a new piece of content on that topic directly to your inbox. The results are new and unique each time, based on the original prompt but never repeating themselves. This makes it a perfect way to passively expand your knowledge without needing to actively search for new content or follow specific sources.

For entrepreneurs, marketers, or business professionals, the tool can be useful too. By setting prompts like “daily updates on AI in marketing” or “daily tips for growing a startup,” you can have the AI fetch fresh tips, insights, and information relevant to your industry. Each result generated by the AI is a fresh piece of advice or knowledge derived from its understanding of the field, offering you something new every day without needing to seek it out manually.

Creative professionals like artists, writers, or designers can also benefit from using the tool to spark new ideas. For instance, you could prompt “daily writing prompts” or “daily color palette inspiration,” and every day, the AI will generate new, unique suggestions or ideas you can build on. Whether you're looking to break creative blocks or just get a fresh perspective, the daily flow of new content can keep your creative juices flowing.

The beauty of this tool lies in its simplicity and automation. You don’t need to continually revisit it to issue new commands or prompts. Once you've entered an initial prompt, the AI runs it automatically every day, generating results based on the prompt through its LLM. The content is always new, whether you're receiving daily quotes, information, or creative suggestions.

Set your initial request and let the results come to you, seamlessly and consistently, every single day.

If you’d like access to the tool just subscribe to my newsletter, and I will send you a link in the welcome email.

Create an ultra-realistic, highly detailed image of a baby boy standing on a sandy beach with a confident and triumphant facial expression. The baby should be around one year old, with short, slightly messy light brown hair and vivid blue eyes looking directly at the viewer. His facial expression should convey a look of determination and confidence, with furrowed brows and a subtle pout with lips pressed together. The baby’s skin is fair, with a natural blush on the cheeks. He is wearing a white t-shirt with bright green sleeves and a green collar. The right hand is clenched into a fist, held close to his chest, with a small amount of sand in it. The focus should be on the baby's upper body, capturing his facial features and the raised fist, while the left hand is not visible. The background is a blurred beach scene with neutral tones to create depth, avoiding distractions. The lighting should be soft, with natural sunlight highlighting the baby's facial details, especially the furrowed brow, giving a sense of depth with gentle shadows. The composition should use a shallow depth of field to keep the baby in sharp focus, with a softly blurred beach background, capturing a triumphant and determined moment.

You will probably get an image that looks like this:

You wrote a prompt about a baby standing on a beach, looking confident and triumphant, and boom—out came something that looks just like the "Success Kid" meme. Pretty cool, but also kind of weird, right? I mean, you didn’t even mention "Success Kid" in your prompt. How does DALL-E / ChatGPT know?

DALL-E isn’t magical (though sometimes it feels that way). It’s actually just really , really good at finding patterns. DALL-E has been trained on a huge number of images from the internet, and a lot of those images are paired with words that describe them. It’s seen millions of pictures, from everything like happy dogs to rainy streets—and of course, it’s also seen famous memes like "Success Kid."

When you give DALL-E a prompt, it looks for everything it knows that matches those words. You mentioned "a baby," "standing on a beach," and "looking triumphant." And here’s the thing: DALL-E has seen that exact combo many times before. It has seen a one-year-old baby on a beach, looking super proud with that tiny fist clenched, and most of the time, that image is tied to one thing: Success Kid. The meme is famous, and DALL-E knows it well.

It’s kind of like when you hear the start of a catchy song. You don’t need the name or all the lyrics to know which song it is. Just a few notes, and you’re singing along. Your brain makes a quick connection because you've heard it so many times before. DALL-E does the same thing, but with pictures. It takes the different things you described—the baby, the beach, the look of victory—and thinks, "Hey, I know what you’re talking about. I’ve seen this before, and it’s that famous meme!"

Another reason this happens is that memes like "Success Kid" are not just famous; they also have a clear, recognizable look. The clenched fist, the confident expression—they all make that image stand out. So when you tell DALL-E you want a baby that looks triumphant, it connects the dots. It finds the closest match to what you want, and, well, there aren’t a lot of images of babies looking victorious on a beach. It figures, "This is the best match," and out comes Success Kid.

What’s super interesting here is that DALL-E works a lot like we do. Our brains are always making connections between things we’ve seen or heard before. We don’t need every detail to understand something—we just need enough clues. When DALL-E generates art, it’s working off clues too, based on the massive number of pictures and descriptions it has seen before. If your prompt lines up with something famous like "Success Kid," it’s going to pull from that, even if you don’t say the exact name.

So, when you use a prompt that sounds a lot like Success Kid without actually saying it, DALL-E gives you Success Kid anyway. It’s not magic; it’s just DALL-E doing what it’s meant to do—finding patterns, connecting the dots, and giving you the best match it can. And since memes are such a big part of what it has learned, it recognizes those famous ones super easily!