General FAQs about our AI Detector

GPTZero is the leading AI detector for checking whether a document was written by a large language model such as ChatGPT. GPTZero detects AI on sentence, paragraph, and document level. Our model was trained on a large, diverse corpus of human-written and AI-generated text, with a focus on English prose. To date, GPTZero has served over 2.5 million users around the world, and works with over 100 organizations in education, hiring, publishing, legal, and more.

Simply paste in the text you want to check, or upload your file, and we'll return an overall detection for your document, as well as sentence-by-sentence highlighting of sentences where we've detected AI. Unlike other detectors, we help you interpret the results with a description of the result, instead of just returning a number.

To get the power of our AI detector for larger texts, or a batch of files, sign up for a free account on our Dashboard.

If you want to run the AI detector as your browse, you can download our Chrome Extension, Origin, which allows you to scan the entire page in one click.

Our users have seen the use of AI-generated text proliferate into education, certification, hiring and recruitment, social writing platforms, disinformation, and beyond. We've created GPTZero as a tool to highlight the possible use of AI in writing text. In particular, we focus on classifying AI use in prose.

Overall, our classifier is intended to be used to flag situations in which a conversation can be started (for example, between educators and students) to drive further inquiry and spread awareness of the risks of using AI in written work.

No, GPTZero works robustly across a range of AI language models, including but not limited to ChatGPT, GPT-4, GPT-3, GPT-2, LLaMA, and AI services based on those models.

• GPTZero is the most accurate AI detector across use-cases, verified by multiple independent sources, including TechCrunch, which called us the best and most reliable AI detector after testing seven others.
• GPTZero builds and constantly improves our own technology. In our competitor analysis, we found that not only does GPTZero perform better, some competitor services are actually just forwarding the outputs of free, open-source models without additional training.
• In contrast to many other models, GPTZero is finetuned for student writing and academic prose. By doing so, we've seen large improvements in accuracies for this use-case.

The nature of AI-generated content is changing constantly. As such, these results should not be used to punish students. We recommend educators to use our behind-the-scene Writing Reports as part of a holistic assessment of student work. There always exist edge cases with both instances where AI is classified as human, and human is classified as AI. Instead, we recommend educators take approaches that give students the opportunity to demonstrate their understanding in a controlled environment and craft assignments that cannot be solved with AI.

The accuracy of our model increases as more text is submitted to the model. As such, the accuracy of the model on the document-level classification will be greater than the accuracy on the paragraph-level, which is greater than the accuracy on the sentence level.

The accuracy of our model also increases for text similar in nature to our dataset. While we train on a highly diverse set of human and AI-generated text, the majority of our dataset is in English prose, written by adults.

Our classifier is not trained to identify AI-generated text after it has been heavily modified after generation (although we estimate this is a minority of the uses for AI-generation at the moment).

Currently, our classifier can sometimes flag other machine-generated or highly procedural text as AI-generated, and as such, should be used on more descriptive portions of text.

1. Help students understand the risks of using AI in their work (to learn more, see this article), and value of learning to express themselves. For example, in real-life, real-time collaboration, pitching, and debate, how does your class improve their ability to communicate when AI is not available?
2. Create an assessment that cannot be answered by Chat GPT or other AI. For example:
   • Ask students to write about personal experiences and how they relate to the text, or reflect on their learning experience in your class.
   • Ask students to critique the default answer given by Chat GPT to your question.
   • Require that students cite real, primary sources of information to back up their specific claims, or ask them to write about recent events.
   • Assess students based on a live discussion with their peers, and use peer assessment tools (such as the one provided by our partner, Peerceptiv).
   • Ask students to complete their assignments in class or in an interactive way, and shift lectures to be take-home.
3. Ask students to produce multiple drafts of their work that they can revise as peers or through the educator, to help students understand that assignments are meant to teach a learning process.
4. Ask students to produce work in a medium that is difficult to generate, such as powerpoint presentations, visual displays, videos, or audio recordings.
5. Set expectations for your students that you will be checking the work through an AI detector like GPTZero, to deter misuse of AI.

Firstly, at GPTZero, we don't believe that any AI detector is perfect. There always exist edge cases with both instances where AI is classified as human, and human is classified as AI. Nonetheless, we recommend that educators can do the following when they get a positive detection:

1. Ask students to demonstrate their understanding in a controlled environment, whether that is through an in-person assessment, or through an editor that can track their edit history (for instance, using our Writing Reports through Google Docs). Check out our list of several recommendations on types of assignments that are difficult to solve with AI.
2. Ask the student if they can produce artifacts of their writing process, whether it is drafts, revision histories, or brainstorming notes. For example, if the editor they used to write the text has an edit history (such as Google Docs), and it was typed out with several edits over a reasonable period of time, it is likely the student work is authentic. You can use GPTZero's Writing Reports to replay the student's writing process, and view signals that indicate the authenticity of the work.
3. See if there is a history of AI-generated text in the student's work. We recommend looking for a long-term pattern of AI use, as opposed to a single instance, in order to determine whether the student is using AI.

We test our models on a never-before-seen set of human and AI articles from a section of our large-scale dataset, in addition to a smaller set of challenging articles that are outside its training distribution. We classify 99% of the human-written articles correctly, and 85% of the AI-generated articles correctly, when we set a threshold of 0.88 on the completely_generated_prob returned by our API (human if below 0.88, AI if above 0.88). Our classifier achieves an AUC score (definition) of 0.98.

To see the full schema and try examples yourself, check out our API documentation.

Our classifier returns a document-level scorecompletely_generated_prob that specifies the probability the entire document was AI-generated. We would recommend using this score when deciding whether or not there is a significant use of AI in generating the text.n our validation dataset, here is how the results change when you set all documents with completely_generated_prob under the threshold as human, and above as AI:

• At a threshold of 0.88, 85% of AI documents are classified as AI, and 99% of human documents are classified as human
• At a threshold of 0.5, 96% of AI documents are classified as AI, and 96% of human documents are classified as human

We recommend using a threshold of 0.88 or higher to minimize the number of false positives, as we think it is currently more harmful to falsely detect human writing as AI than vice versa.

Additionally, we highlight sentences that been detected to be written by AI. API users can access this highlighting through the highlight_sentence_for_ai field. The sentence-level classification should not be soley used to indicate that an essay contains AI (such as ChatGPT plagiarism). Rather, when a document gets a MIXED or AI_ONLY classification, the highlighted sentence will indicate where in the document we believe this occurred.

No. We do not store or collect the documents passed into any calls to our API. We wanted to be overly cautious on the side of storing data from any organizations using our API.

However, we do store inputs from calls made from our dashboard. This data is only used in aggregate by GPTZero to further improve the service for our users. You can refer to our privacy policy for more details.

You can use the following bibtex citation:

@misc{tian2023gptzero,
    publisher = {GPTZero},
    url       = {https://gptzero.me},
    year      = {2023},
    author    = {Tian, Edward and Cui, Alexander},
    title     = {GPTZero: Towards detection of AI-generated text using zero-shot and supervised methods}
}