TextFromImage — Privacy policy

Last updated: May 2026

Summary: OCR runs entirely on your device. AI4Context does not operate a backend for this extension and does not receive your images or extracted text.

What is processed

When you upload, paste, drag-and-drop, or select an image from a web page, the extension runs local OCR (Tesseract.js) in your browser. Extracted text stays on your device unless you copy or download it yourself.

Web images

If you use the context menu on a page image, the extension may request permission for that site’s origin to fetch the image bytes. Those bytes are processed locally. No image or OCR result is sent to AI4Context servers.

What we store locally

• UI language preference (Spanish / English).
• Optional panel layout preferences (preview and result area heights).
• Temporary session data when opening an image from the context menu (cleared when you use the panel).

Permissions

• storage — preferences and temporary image queue between the background worker and panel.
• contextMenus — “Extract text from image (TextFromImage)” on images.
• windows — open or focus the floating OCR panel.
• optional host permissions — fetch an image from a website only when you use the context menu on that site (granted per origin).

Third parties

This extension does not call cloud OCR APIs. Tesseract language data is bundled in the extension package (English and Spanish).

Contact

AI4Context — ai4context.com