AI Detector Pro is a tool that tells you if text was written by a human or AI (GPT or BARD based).
AI Detector Pro detects text content generated by GPT and BARD based AI algorithms. These are used in almost all AI article generation tools. They also are used in popular grammar and style correction tools. AI Detector Pro detects text created by any of these tools.
Email us at firstname.lastname@example.org -- we're happy to help.
We offer a money back guarantee which you can read about here. For any questions on the guarantee, please contact us.
Our AI detection algorithms are very compute-intensive. They need to be run on GPUs optimized for AI applications, which are extremely expensive still. For this reason, each report costs us a non-insignificant amount of money.
No. Language generation models are based on algorithms that have a random component. There is no way to be 100% accurate in identifying these. In the vast majority of cases, AI Detector Pro is accurate, and we are constantly improving and updating our algorithms to maximize accuracy.
Currently, AI Detector Pro only works in English. Our system was trained on an English-language dataset, so it's not able to properly classify other languages. However, we are looking into expanding it in the future to work with other languages.
AI Detector Pro uses a large amount of data produced by GPT based AI text generation models to find patterns in the data. These patterns are much more likely to occur in text written by AI than in text written by a human. The more of these patterns there are in an article, the more likely it was to have been generated by AI.
The AI Eraser helps you humanize your content by finding words and phrases that trigger AI detection tools. You can the content manually to remove the AI, or you can use AI Eraser's suggestions to speed up the process. Since AI Eraser's suggestions are not based on GPT or BARD, they're less likely to trigger AI detection tools. AI Eraser scans each suggestion and ranks it by likelihood of detection.
AI text generation is based on a modified neural network called a "transformer" network. In training the network, the computer reads a huge dataset and looks at relative probabilities of words. For example, the word "is" comes after the word "he" very often, so the computer learns that when it writes text, it should follow "he" with "is".
This is complicated by something called "attention", which allows the computer to look at other words in context to determine which words should be generated. For example, the computer will often see the name "Mark" and then later "he" in a sentence; however, it will rarely see "Mark" and then "she" in a sentence. Because of this, when asked to generate a story about "Mark", it will use "he" instead of "she".
The free version includes 3 reports. After that, you can upgrade to a paid plan.
Each time AI Detector Pro scans a document, it counts as one report. If you scan the same document multiple times, it counts as multiple report. If you use the AI Eraser, each time you click "Scan" you're running a new report.
The free version includes a some AI Eraser functionality; however, it doesn't provide automated suggestions.
Many grammar checkers and writing tools use GPT under the hood. Because of this, if you used a popular grammar/writing tool, AI Detector Pro (and other AI detection tools) may detect it as AI.
You can cancel by logging in and going to your Account page from the bottom left of the screen. From there, you can click to cancel.
AI Detector Pro has been trained on GPT based data (GPT-2, GPT-3, GPT-4, GPT-J), as well as BARD. Note that the vast majority of online article generation tools or article rewriting tools are based on GPT. GPT is the state-of-the-art and no other known models come close to its accuracy.