Privacy Policy

Last updated: 2026-09-12

PDF data

PDF splitting is designed to run in your browser. PDF contents are not uploaded to the Split2AI application server. Filenames and page content are used only locally for the split workflow.

Analytics

Split2AI may use Vercel Web Analytics for aggregate usage data such as page views. PDF content, page text, filenames, exact file sizes and exact local processing times are not sent to analytics.

Optional anonymous feedback

After splitting, feedback is sent only when you explicitly press a success or failure button. It contains only the selected AI service, success/failure outcome and a coarse output-size band. It does not contain PDF content, filenames, page text or exact file size. Observation URLs accept only allowlisted service/outcome values and their responses are configured not to be stored in caches.

Referrers and device permissions

Production responses use a no-referrer policy. Split2AI also disables browser permission access for camera, microphone and geolocation because those capabilities are not required for PDF splitting.

Third-party libraries

On first load, libraries such as pdf-lib and JSZip may be fetched from a CDN, which receives normal web request metadata.

Affiliate links

If affiliate links are introduced, they will be disclosed. PDF content, page text and filenames will not be provided to affiliate partners.

Split2AI · Large PDFs, prepared for AI.