Skip to main content
Strict Privacy Architecture

Privacy Policy

Learn how Compixor AI processes files, handles technical information, and manages data when you use our online PDF and document tools.

Last Updated: November 2026 • Effective Immediately

The Compixor AI Privacy Commitment

This Compixor AI Privacy Policy outlines our transparent approach to file privacy and security. We maintain an architectural distinction between client-side and server-assisted operations: nine tools process documents completely inside your local browser memory, while three specialized conversion utilities use secure, temporary server processing over encrypted HTTPS. We do not store your files, inspect their contents, or retain facial biometric data. Standard technical server logs are maintained strictly for performance, uptime, and security diagnostics.

Privacy at Compixor AI

At Compixor AI, we believe user privacy should be built into system architecture rather than treated as an afterthought. When you use our platform for online PDF tools privacy, you are never required to register an account, submit an email address, or enter personal identity credentials.

Our core mission is to provide high-performance document and media utilities with minimal data exposure. By leveraging modern web standards, we avoid centralized file storage and reduce data transmission to the absolute operational minimum required for each specific tool. For further context regarding our platform mission, visit our About Us page.

Local and Server-Side File Processing

To deliver both absolute privacy and advanced conversion capabilities, our PDF and document file processing is structured into two clearly differentiated operational models:

Client-Side Browser Processing (9 Tools)

For nine of our utilities, browser-based PDF processing runs entirely on your local device. The application code executes in your browser memory utilizing WebAssembly (Wasm), Web Workers, and HTML5 Canvas APIs. Your file bytes are never uploaded to our servers or transmitted over the internet.

Server-Assisted Conversion & Processing (3 Tools)

Certain specialized tasks require advanced computing environments and machine learning models that cannot run efficiently across all mobile devices or browser memory limits. Three tools utilize secure server-assisted processing:

  • Biometric Passport Photo Maker: Your photo is sent securely to our server only to remove the background. We do not store it. Your edited photo stays in your browser until you download it or close the page.
  • Image to Word: Transmits the image over encrypted HTTPS to our processing backend for multi-lingual optical character recognition (OCR) and DOCX document reconstruction.
  • PDF to Word: Transmits the source document over encrypted HTTPS for layout analysis, text extraction, font preservation, and conversion into editable Microsoft Word (.docx) format.

Secure Ephemeral Lifecycle: For these three server-assisted utilities, data is processed over encrypted HTTPS and not stored; deleted right after processing once conversion finishes and the file is returned. Files are never retained on our servers.

Information We Collect and Technical Logs

We adhere strictly to data minimization principles. We distinguish between operational technical metadata and user file content:

Technical Logs Collected

When you connect to our website and APIs, standard automated web server logs record technical diagnostics:

  • Internet Protocol (IP) address
  • HTTP request method and destination path
  • HTTP response status code
  • Browser user agent string
  • Timestamp and processing latency

These logs are maintained solely for network security, DDoS mitigation, and server performance monitoring.

Information We Never Collect

We design our systems to ensure your document data remains completely confidential:

  • No document contents, text excerpts, or metadata indexing
  • No facial embeddings, facial geometry, or biometric templates
  • No passwords or decryption keys entered in PDF tools
  • No QR code text, URL destinations, or payload data
  • No personal user profiles or marketing tracking IDs

Browser Storage and Cookies

Compixor AI does not utilize third-party advertising cookies, cross-site tracking beacons, or commercial profiling scripts.

We utilize lightweight, client-side browser localStorage strictly to store functional preferences:

  • Theme Preference: The compixor-theme key remembers whether you prefer dark or light mode interface styling across browser visits.
  • Functional Counters: An internal counter stores the number of files processed on your current device to trigger non-intrusive user feedback prompts.

To understand overall site performance and uptime, we utilize Vercel Analytics, a privacy-first web telemetry service that records aggregate, cookieless metrics (such as page views and general device categories) without capturing personal identity or tracking individuals across the web.

Third-Party Processing Services

We are transparent regarding any third-party file processing required to deliver advanced functionality:

Google Gemini AI (Urdu and Arabic OCR)

When using Image to Word or PDF to Word for complex Urdu or Arabic document transcription, designated image segments may be sent to Google’s Gemini API as an external processor. These transmissions occur strictly for character transcription and text extraction during conversion. Transmitted data is not used to train proprietary public models.

Our website infrastructure and content delivery are hosted through enterprise cloud providers (including Vercel) that maintain SOC 2 compliance and rigorous transport-layer encryption.

File Retention and Data Security

Our document processing and data retention model is governed by zero permanent storage:

  • Zero File Retention: We do not retain, archive, or backup user documents. Once a file conversion finishes and the result is transferred to your device, the temporary server buffer is immediately erased from system storage.
  • Transport-Level Encryption: All communications between your web browser and our processing backend are encrypted using modern Transport Layer Security (HTTPS / TLS 1.3).
  • No Model Training: User document contents, uploaded photos, and transformed text are never used, analyzed, or sold to train artificial intelligence or machine learning models.
  • No Facial Data Retention: When processing biometric passport photos, facial landmarks and cropping dimensions are discarded immediately upon file download.

Privacy Rights and Contact Information

Compixor AI designs its tools around data minimization and privacy-first principles. Because we do not require account sign-ups, maintain user profiles, or store uploaded files, we generally do not hold personal data linked to your identity.

If you have questions regarding this Privacy Policy, technical logs, or how our tools process data, please contact our support team:

Changes to This Privacy Policy

We may periodically revise this Privacy Policy to reflect technical updates, infrastructure improvements, or new tools added to our catalog. Any modifications will be posted directly on this page with an updated revision date. We encourage users to review this page periodically to remain informed about how we safeguard file privacy.