What it is
Overview
An AI detection tool that claims to identify machine-generated text across 26+ models including ChatGPT, Claude, and Gemini. Pangram uses a proprietary detection system rather than relying on third-party APIs. The typical user base includes academic administrators and content moderators managing submissions at scale, drawn by the Google Docs integration and batch processing features that let them check multiple documents without manual uploads.
At a glance
Usability & Quality overview
Best for
- educators
- teams needing quick AI-content checks
- users who want Google Docs/browser-based workflows
Watch out for
- false positives on essays and research writing
- small and noisy public review sample
- limited evidence from independent long-form user posts
Real product, not a wrapperIndependent product
Pangram has developed proprietary AI detection models specifically trained to identify content from 26+ different AI models. This represents genuine specialized technology rather than a simple wrapper around existing detection APIs.
Strong evidenceQuality score
Pangram AI text detection with browser and Google Docs workflows, but false positives on human writing.
This score is our editorial judgment, computed automatically from the sources, weights, and dates shown above. It reflects the data we could verify as of August 18, 2026, not a guarantee or statement of fact about Pangram. Third-party ratings and quotes belong to their original platforms and authors. Thin data lowers our confidence label, and we say so instead of guessing. Work on Pangram? Dispute any datapoint and we will review it, publish your response, and correct verified errors.
Plans
Pricing
How free is free?
Free with limits
2,000 words/day + 3 image scans; Individual $20/mo for 300k words
What you get for free
- 2,000 words per day detection limit
- 3 image detection scans daily
- AI assistance detection included
- Interpretability features
- File upload and OCR support
- Detect AI-generated and human content in 20+ languages
- Browser extension access
- Google Docs integration
Behind the paywall
- Plagiarism detection on every scanIndividual
- Higher word limits (300k+ words/month)Individual
- More image detection scans (100+ monthly)Individual
- Automatic social media and Substack feed scanningIndividual
- API usage credits ($200+ monthly)Professional
Cost to actually use
Based on 20 classified review complaints about rate limits, credits, and billing.
Community feedback
Aggregated reviews
Ratings and quoted comments below are aggregated from third-party sources and reflect those users' views, not SearchTools.ai's.
What reviewers talk about
themes inside the Sentiment pillar — not score ingredients
25Output Quality117 mentions
“As most of you here, I actually found Pangram to be quite reliable. I teach MBA students in India and usually build and test many tools to enhance my writing, teaching and grading. Although I have nothing against AI, as business students need to be on top of whatever tech is coming, I hate it when they use AI tools to write up their case reports. I used Pangram after extensively testing it on my own writing. I found it best because it was always very conservative in its flagging. This made it pa”
“I wrote this essay myself, about my life and English is my second language. AI corrected my grammar, that is all. Pangram told me it was '100% AI, the entire text.' That is seriously BS. Human-written with grammar help is not '100% AI,' and a tool that cannot tell the difference has no business being that confident about a person's work. Thank you for ruining my valuable work that was called out by idiot people who deeply trusted in your silly AI who cannot tell the difference. SHAME ON YOU!!! ”
“This is not consistent and unreliable. This flagged my cover pages/title pages/index at 88% AI, and marked my reference page as being plagiarism. Also indicated a literature review that I did, which has in-text citations throughout, as AI. It's cited because they aren't my ideas/words; credit was given to the original author, and it is not AI. This also flagged the data I collected from a survey as being AI, when it is actual data points that I received, and I attached an image as an appendix”
“So either I'm the luckiest person alive or Pangram's 99.98% claims of accuracy are a bit inflated. I don't know what the point of a tool that just highlights all your text and says it's 100% AI without telling you how to fix it or pointing to particular sentences. It's almost like using a tool to feel terrible about yourself. Everything I write gets flagged unless I change my writing style to deliberately sound unintelligible. So I guess I'm either in the 0.02% repeatedly or I'm so lucky I'm goi”
31Value & Pricingthin data · 8 mentions
“I tried the product out for a week, I liked it but it wasn't the right time for a subscription. I mistakenly let the first billing date pass and got charged (my fault). James from their team was helpful in getting me a full refund and cancelling the paid account without any hassles. Great service!”
“Can't even get my 7-day free trial without putting my card info in.”
“gross and weird. I have never known a tech company not to attempt to profit off any data they can extract from customers, so like…. pretty sure they’re going to use our writing to train LLMs or something. They’re already using Reddit and Substack would be a great medium between academia/literature-level writing and how people talk IRL.”
“It also doesn't scan anything uploaded prior to 7/21, so I'm wondering if it gets stored in a data center somewhere upon upload.”
21Reliability13 mentions
“As most of you here, I actually found Pangram to be quite reliable. I teach MBA students in India and usually build and test many tools to enhance my writing, teaching and grading. Although I have nothing against AI, as business students need to be on top of whatever tech is coming, I hate it when they use AI tools to write up their case reports. I used Pangram after extensively testing it on my own writing. I found it best because it was always very conservative in its flagging. This made it pa”
“This is not consistent and unreliable. This flagged my cover pages/title pages/index at 88% AI, and marked my reference page as being plagiarism. Also indicated a literature review that I did, which has in-text citations throughout, as AI. It's cited because they aren't my ideas/words; credit was given to the original author, and it is not AI. This also flagged the data I collected from a survey as being AI, when it is actual data points that I received, and I attached an image as an appendix”
“Ever since I wrote my last review. 12 days ago. Not even one paper that this company has not flagged with AI. Says all papers are 100% AI. This includes sections like the cover page and references. PANGRAM SHOULD NOT BE USED AS AN AI DETECTION TOOL. IT KNOWS NOT WHETHER AI WAS USED OR NOT.”
“routinely flags both published, pre-LLM human-authored texts, and texts developed by intensive model co-editing processes as "100% AI". this has increased greatly since Pangram 4 rolled out. ”
55Ease of Use13 mentions
“I'm on the trial. I like the interface. All the students I suspected of using AI were positive for it. I tested Pangram a bit by submitting 8 sample sections of my own notes. All came back human with high confidence. That was reasssuring. Could have a Westworld situation. I also submitted some (5 samples) of AI produced work. All came back as AI with high confidence. That's not a big enough test to be very confident that it's accurate, of course. But it counts in favor of it. I plan to keep u”
“So either I'm the luckiest person alive or Pangram's 99.98% claims of accuracy are a bit inflated. I don't know what the point of a tool that just highlights all your text and says it's 100% AI without telling you how to fix it or pointing to particular sentences. It's almost like using a tool to feel terrible about yourself. Everything I write gets flagged unless I change my writing style to deliberately sound unintelligible. So I guess I'm either in the 0.02% repeatedly or I'm so lucky I'm goi”
“Can't even get my 7-day free trial without putting my card info in.”
“Great review performance, but still cumbersome interface and capacity to streamline results. Also unable to handle large files with a lot of images”
30Trust derived from dimensions + predator detectionview math
A composite of the quality dimensions weighted by mention volume, then capped by predator / abuse-detection rules.
Reasoning
Watch & learn
Video content
YouTube
YOUTUBE1.6K viewsAI Detection is a Scam
WordslingerPodcast26 days ago
YOUTUBE1.3K viewsBest AI humaniser Versus TOP AI Detectors | July 2026 Update
Insights4UToday1 month ago
YOUTUBE167 viewsWhy I use Pangram 4 (and why you shouldn't trust it blindly)
brnnnmcdnld19 days ago
YOUTUBE320 viewsAI Detection is ANTI-HUMAN
FutureFictionAcademy26 days ago
YOUTUBE118 viewsCan AI Detectors Be Trusted? Pangram 4 and the False-Positive Problem
DrLitman19 days ago
YOUTUBE204 viewsSubstack Partners with Pangram to Scan For AI Assisted Content... But How Much AI Is Too Much?
ingridzapataread26 days ago
Capabilities
Key features
Content Detector
Analyzes text to estimate whether it was written by a human or AI
Plagiarism Checker
Scans text against sources to detect copied passages and flag originality issues
Detector
Identifies whether content was AI-generated by analyzing text or media signals
The honest take
What users love & flag
Distinct themes surfaced across 21 reviews from 2 sources — each grounded in real review text, ranked by how often it comes up.
Questions
Frequently asked
What is Pangram?
Pangram is an AI content detection tool that identifies AI-generated text and images with 99.98% accuracy. It can detect content from over 26 AI models including ChatGPT, Claude Opus 5, Gemini 3, and others, while also providing plagiarism checking and supporting over 20 languages.
Is Pangram free to use?
Yes, Pangram offers a free tier that includes 2,000 words per day and 3 image detection scans daily, plus browser extension access and Google Docs integration. Paid plans start at $20/month for the Individual plan with 300,000 words and 100 image scans monthly.
How accurate is Pangram compared to other AI detectors?
Pangram achieves a false positive rate of just 1 in 10,000 (0.01%), significantly outperforming competitors like GPTZero which has a 2.4% false positive rate. This accuracy has been verified by independent research from the University of Chicago and University of Maryland.
Can Pangram detect AI-edited human text?
Yes, Pangram can detect AI assistance in human-written text that has been edited or enhanced by AI tools. This feature helps identify content that may have started as human writing but was subsequently modified using AI assistance.
What file formats does Pangram support?
Pangram supports multiple input methods including copy-paste text, file uploads in PDF, DOCX, and RTF formats, plus browser extensions and Google Docs integration. It also includes OCR capabilities for processing scanned documents.
Does Pangram work with learning management systems?
Yes, Pangram offers enterprise solutions with custom LMS integrations for Canvas, Google Classroom, Moodle, and Brightspace. These integrations allow educators to seamlessly incorporate AI detection into their existing workflows and grading systems.
Can Pangram detect AI-generated images?
Yes, Pangram includes image detection capabilities that can identify AI-generated visual content, currently available as a research preview feature. The free plan includes 3 image scans daily, while paid plans offer 100-500 monthly image scans.
What languages does Pangram support for AI detection?
Pangram supports AI detection across 20+ languages, making it suitable for international educational institutions and global enterprises that need to verify content authenticity in multiple languages.
Does Pangram show which specific parts of text are AI-generated?
Yes, Pangram provides segment-by-segment analysis that highlights specific portions of text identified as AI-generated and indicates the degree of AI involvement. It also includes interpretability features that explain which text patterns triggered the AI detection.
Compare Pangram
Compare with another toolMore Like This
