Human, AI, and Mixed estimates for arXiv abstracts. Powered by GPTZero. Abstracts only; no API key needed.
Users
45
+40 in 30 days
Rating
5.0 / 5
unchanged in 30 days
Ratings
2
0 in 30 days
Store rank
#114K
#95K among extensions
Version
3.1.7
updated 13 Oct 2026
Users
monthly, 1 months
Latest, 18 Oct 2026:45 usersMove the mouse over the chart for any month
Ratings count
monthly, 1 months
Latest, 18 Oct 2026:2 ratingsMove the mouse over the chart for any month
The Chrome Web Store rounds user counts down to the leading digit, so the user line moves in steps. The ratings count is exact and is the better signal of real adoption.
About Abstract Lens
Read arXiv’s new-submission pages with labels for AI and human-written abstracts.
Abstract Lens adds a small colored tag to each submission: "Human" (green), "AI" (red), or "Mixed" (yellow). This comes from GPTZero’s analysis of the abstract. Click on a tag to see classification probabilities and confidence.
# Notes
• Just abstracts: no PDFs or full manuscripts are analyzed.
• No personal account needed.
• Works on arXiv new-submission pages in Chrome and Brave.
# About Permissions
• Permissions limited to arxiv.org new-submission pages (the narrowest possible to make this extension work).
• Chrome asks for permissions to “read and change your data on arxiv.org” so it can add the classification tags to the page. It does not modify anything else on this or any other page.
An automated estimate is not proof of AI or human authorship. Labels are generated only from the abstract, not the full paper. No judgement is made on scientific quality of any paper. Abstract Lens is independently developed. It is not affiliated with or endorsed by arXiv or GPTZero.