Professional AI Transcription with Zero Compromise on Privacy.

Samizdat AI is a local-first, open-source toolkit for journalists, researchers, and legal professionals. Your sensitive data is processed on your machine, never in the cloud.

Join the waitlist for the V1 beta. No spam, ever.

The Pillars of Trust

Local-First & Secure

Your data is processed on your device. We never see it. This "digital shield" architecture protects you from data breaches and third-party subpoenas.

Transparent & Open Source

Our code is public and auditable on GitHub. We believe in verifiable trust, not blind faith. No black boxes.

Purpose-Built for Professionals

A workflow designed for the rigor of reporting, research, and legal work—not for generic business meetings.

Our Vision

PHASE 1 (LAUNCHING SOON)

Master Your Audio

Our first release is relentlessly focused on perfecting a single, critical workflow: providing a flawless, private, and trustworthy transcription experience. This is the bedrock of trust upon which everything else will be built.

THE NEXT HORIZON

Unify Your Research

We know crucial information isn't just in audio; it's in scanned documents, leaked PDFs, and photos from the field. Our next major step will be to extend Samizdat AI's privacy-first promise to this world, creating a single, searchable, and secure local archive of all your text-based source material.

THE ULTIMATE GOAL

Connect the Dots

The future of Samizdat AI is to become your "Storycraft OS" for investigative work. Imagine instantly seeing every time a person mentioned in an interview also appears in a thousand-page document dump—all locally, all privately, all under your control.