We never share or promote your knowledge to any third parties. Sure, all person information Pangram AI Detector is encrypted and stored securely on our servers to be displayed in your historical past. Extra credit and features like plagiarism detection can be found with a subscription. This common retraining process ensures our detection capabilities keep present with evolving AI applied sciences and might reliably establish content material from the latest language models.
The authors reported a decline after 2022 within the share of abstracts with low AI scores. A 2026 analysis for Organization Science used Pangram to look at submitted abstracts and peer evaluations. Pangram categorised about 41% of the LinkedIn posts longer than 250 words in that pattern as AI-generated. Pangram flagged about 9% of the articles as partially or wholly AI-generated. In March 2026, NewsGuard launched a software using Pangram to establish web sites with substantial quantities of AI-generated content. Pangram is out there on the corporate’s web site, via browser extensions and thru an API.
Tools that claim to catch ChatGPT or Claude output have multiplied rapidly — and most of them rely on the identical flawed statistical shortcuts. Pangram is significantly harder to bypass than most AI detectors as a result of its training explicitly accounts for evasion techniques. Pangram was particularly skilled with adversarial examples to detect AI content that has passed via paraphrasing instruments like QuillBot, or dedicated AI humanizers. Entry per-sentence confidence scores, mannequin attribution signals, and a shareable PDF report — ideal for tutorial or professional documentation. Copy any doc, essay, article, or email into the Pangram AI Detector input field. Pangram’s accuracy isn’t a advertising declare — it’s a peer-reviewed measurement.
Accuracy on AI-edited textual content (where a human wrote the majority and used AI for refinement) drops to approximately 73% — the identical limitation that affects all current detectors. This matters as a result of most academic insurance policies distinguish between using AI to shine grammar versus utilizing it to put in writing the total document. Of all business and open-source detectors examined, only Pangram met this threshold while nonetheless sustaining dependable detection accuracy. The study used a managed corpus of pre-2020 human texts alongside AI-generated content from four frontier fashions, stress-tested across a number of genres, text lengths, and evasion conditions.
The enter is handed by way of the neural network, producing an output vector. Our classifier uses a traditional language mannequin architecture. This is why we’re delighted to be partnering with Pangram, which has set the gold commonplace for Al detection and attribution Pangram is actually one of the best AI detectors we’ve, and y’all should use it. Amongst automatic detectors, Pangram significantly outperforms the remainder. We examine human specialists with five AI detectors, together with Pangram and GPTZero.
Teachers And Professors
Most AI detectors measure perplexity — a statistical measure of how “surprising” each word is given what got here earlier than. Paraphrasers, humanizers, and translation tips all cut back effectiveness by roughly 2-3%, in comparison with 40-60% drops seen in simpler detectors. It additionally detects text the place evasion was attempted by way of machine translation. Its detection rate on paraphrased AI content is approximately 97%, compared to 40-60% for most standard detectors. The false positive price — meaning human textual content incorrectly flagged as AI — is approximately 1 in 10,000, the lowest independently audited figure in the industry. Pangram’s sentence-level reviews can help proof disclosure and inner audit processes.
Unbiased Evaluations
- Pangram is over 99% correct and has been third-party verified by researchers on the University of Chicago and The University of Maryland.
- Its detection price on paraphrased AI content is approximately 97%, compared to 40-60% for most standard detectors.
- Pangram identifies these micro-signatures across the total doc.
- Our browser extension is a fast and handy means to make use of our AI detection tool whereas shopping the web or inside web-based purposes corresponding to Google Docs.
Pangram’s false optimistic price (the speed at which human documents are incorrectly flagged as AI) is at present 1 in 10,000. Our accuracy and reliability have been independently verified by third-party researchers and academic establishments. Pangram depends on various datasets, onerous unfavorable mining, and energetic learning to achieve industry-leading FPRs rather than perplexity and burstiness, which frequently fail in apply.
In tests by Pangram reported in July 2026, Pangram four incorrectly flagged 41 of 1,000,000 human-written English samples from FineWeb. The detectors usually categorized minimally edited text as AI-generated and struggled to distinguish levels of AI assistance. When the members’ judgments were combined by majority vote, they appropriately categorized all but considered one of 300 English nonfiction articles. Logographic scripts, or writing methods corresponding to Chinese Language that do not use an alphabet however are composed principally of logograms, cannot produce pangrams in a literal sense (or no less than, not pangrams of affordable dimension). This differs farther from English pangrams, with letters such as ç, ä, and š.
Paste Your Textual Content
When the stakes embrace litigation, use a detector with documented accuracy. Confirm scholar submissions with a tool your educational integrity board will accept. From K-12 teachers to enterprise publishers, Pangram AI Detector adapts to the stakes of the work.
Our browser extension is a fast and convenient way to make use of our AI detection tool whereas browsing the online or inside web-based applications corresponding to Google Docs. Pangram’s LMS integrations permit universities to create knowledgeable AI insurance policies. Pangram is also integrated into Chrome browser, Google Docs, and plenty of learning management techniques like Google Classroom and Canvas.
Pangram analyzes your text using advanced AI detection algorithms to establish AI-generated content material with excessive accuracy. Copy and paste your textual content into the sphere or upload recordsdata. Based in Brooklyn, NY, in 2023 by AI researchers from Tesla and Google, we proceed to contribute to the analysis neighborhood to extend transparency. Our AI detection works throughout greater than 20 languages, making it a truly international, multilingual solution for establishments and companies worldwide. Pangram can detect content material from all major language models together with ChatGPT, Claude, Gemini, Llama and more, making it a comprehensive answer for AI detection. We additionally present superior features backed by peer-reviewed research that enable users to understand the origin of an article.
Pangram presents a free detection check with prompt results — paste your textual content and receive your AI rating with no login required. It analyzes structural, stylistic, and semantic patterns in textual content — including word-choice tendencies, sentence transition patterns, and generation artifacts particular to large language fashions. Translation-based evasion — where AI text is translated to another language and back — can also be caught.
In a 2025 examine, Jenna Russell, Marzena Karpinska and Mohit Iyyer compared automated detectors with individuals skilled in utilizing language models for writing. For Pangram four, the corporate reported a false constructive price of zero.0041%, roughly one in 24,000, on an English-language benchmark. Whereas the English language makes use of all 26 letters of the Latin script in native and naturalized words, many different languages utilizing the same alphabet don’t.
