# For Developers # Integrate with the best AI Detection API on the planet. #### Get industry leading AI insights and statistics behind every article, with the same API used by GPTZero to serve 8M+ users. [View docs](https://gptzero.stoplight.io/) [Get API key](https://app.gptzero.me/api) ```shell curl --request POST --url https://api.gptzero.me/v2/predict/text --header 'Accept: application/json' --header 'Content-Type: application/json' --header 'x-api-key: ' --data '{ "document": "string", "version": "string" }' ``` # Pricing Choose the right plan for your needs. Starting at $45/month. 1m words each month $135 /month 2m words each month $250 /month 300k words each month $45 /month 10m words each month $1000 /month 20m words each month $1850 /month 5m words each month $550 /month * * * Estimate cost after additional usage After the base plan word allotment is used, additional usage is charged at $150 per million words. Total words Additional words: 0 Base plan cost: $135.00 Additional usage cost: $0.00 * * * Estimated total monthly cost: $135.00 Selected plan 1m words each month $135 /month 50 file batch uploads 150k characters per doc Access to GPTZero’s premium detection model Get started # Ecosystem _Translate_ uses the GPTZero API to ensure authentic translations. [Learn more](https://www.translate.com/) _K16 Solutions_ uses GPTZero to power AI detection analytics for colleges. [Learn more](https://www.k16solutions.com/product/scaffold-ai-detection/) _Nextcloud_ uses GPTZero to helps organizations regain control over their data. [Learn more](https://nextcloud.com/blog/introducing-hub-5-first-to-deliver-self-hosted-ai-powered-digital-workspace/) _Newsguard_ leverages GPTZero to protect the integrity of the internet [Learn more](https://gptzero.me/news/newsguards-gptzero-powered-investigation) _Microsoft Threat Intelligence Center_ uses GPTZero to monitor AI-driven misinformation. _Schmidt Futures_ uses GPTZero to understand AI trends in scholarship applications. _Kritik_ leverages GPTZero to deliver quality learning experiences to 250+ academic institutions. [Learn more](https://help.kritik.io/en/articles/8011603-ai-generated-content-check) _Source Scout_ uses GPTZero to help journalists vet new sources. [Learn more](https://aliciastewart.com/) _Peerceptiv_ uses GPTZero to support trusted peer to peer learning communities. [Learn more](https://peerceptiv.com/resources/gptzero-and-peerceptiv-agree-to-test-ai-detection-in-peer-learning/) _Brisk Teaching_ uses the GPTZero API to power a growing teacher's app. _Revisely_ uses GPTZero to power a leading education platform. [Learn more](https://revise.ly/revisely-integrates-gptzero-to-detect-ai-generated-text-in-student-documents/) _Coding it Forward_ uses GPTZero to ensure integrity in applications. [Learn more](https://gptzero.me/news/how-one-non-profit-dealt-unexpected-and-ai-generated-content-and-turned-to-gptzero) _Coursify Studios_ is leveraging GPTZero to build the next generation LMS. ### Compliance GPTZero is proud to be AICPA SOC 2 compliant, ensuring excellence and trust in every cutting edge product we bring our users. SOC 2 is a voluntary standard by the American Institute of CPAs for service organizations concerning how to handle customer data. It's based on five criteria: security, availability, processing integrity, confidentiality, and privacy. ## FAQs about GPTZero Everything you need to know about GPTZero and our ChatGPT detector. ### How does GPTZero detect AI-generated text? GPTZero uses deep learning models that keep pace with AI advancements to deliver precise, reliable results that help you understand and interpret the origin of a piece of text. A sentence-by-sentence classifier calculates the probability that a text was created by AI, giving you both document-level and granular insights. ### How accurate is GPTZero at detecting AI text? Independent and internal benchmarking show that GPTZero outperforms competitors at detecting mixed documents (where both AI and human writing has been used) with 96.5% accuracy. Our false positive rate is under 1%, making it one of the most reliable detectors available. ### Can GPTZero detect text from the latest AI models? Yes. GPTZero is continually updated to recognize text from the newest large language models, including ChatGPT (GPT-3, GPT-4, GPT-5), Google Gemini, LLaMA, Claude, and others. As new models are released, we adapt and update our detection system. ### How does GPTZero reduce false positives, especially for ESL students? Since 2022, we've focused on reducing bias for ESL (English as a Second Language) writers. By tagging educational data, adding representative datasets, and using text pre-classification, we've reduced the false positive rate on TOEFL essays to just 1.1%. ### How does GPTZero handle paraphrased or modified AI content? Many try to bypass AI detectors by paraphrasing or using homoglyph substitutions. Our model's Paraphraser Shield technology is designed to catch these cases. Even if AI content has been altered to look more human-like, GPTZero can detect it. ### What are the known limitations of AI detection? No AI detector is 100% accurate, and AI itself is changing constantly. GPTZero performs best on longer texts and English prose. We encourage using it as a conversation starter, and not as the final verdict. GPTZero is the only detector specifically de-biased for ESL writers, lowering false positives to 1%. ### What data did you train your model on? Our model is trained on millions of documents spanning various domains of writing including creative writing, scientific writing, blogs, news articles, and more. 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. ### How do I use and interpret the results from your API? When you run text through our API, you'll see a `document_classification` field that labels it as `HUMAN_ONLY`, `MIXED`, and `AI_ONLY`. Each classification comes with probabilities (via the `class_probabilities` field) and a `confidence_category` field, which can be high, medium, or low. When the confidence is "high", error rates are below 1%. The API provides sentence-level highlights (API users can access this highlighting through the `highlight_sentence_for_ai` field) so you can see exactly which parts of the text contributed most to the classification. ### Are you storing data from API calls? No. We do not store or collect the documents passed into any calls to our API. For dashboard users, only aggregate inputs are stored to improve the service. See our [privacy policy](https://gptzero.me/privacy-policy) for full details.