Security by architecture, not afterthought
Most apps add security around the edges. Mizan is built the other way: the protections below are part of the core engine, not a layer on top.
Encryption everywhere
- AES-256-GCM encryption is applied per provider to every connection token — the keys that link your accounts are never stored in plain text.
- TLS protects all data in transit, and data is encrypted at rest by our infrastructure providers.
A ledger you can prove
Every figure in Mizan is written to an immutable, hash-chained audit ledger. Each entry is cryptographically linked to the one before it, so your numbers are tamper-evident and verifiable a year later — not a snapshot that silently changes.
A native app, with bounded AI
Mizan runs as a native desktop application backed by a Rust-native financial-truth engine — not a browser tab holding your credentials. The AI layer can read everything it needs, but it can only write back to your ledger through an audited, bounded path. It can suggest; it cannot silently act.
We see less than you'd think
Account connections are designed for read access wherever possible, and sensitive bank credentials are handled by regulated aggregators rather than stored by us. We practise data minimisation — we keep what's needed to give you an accurate picture, and no more.
The waitlist, today
Right now, before launch, the only data we hold is your waitlist email and country. It lives in a secured database with row-level security and is never sold. See our Privacy Policy for the full detail.
Responsible disclosure
If you believe you've found a security vulnerability, please tell us first at info@getmizan.net. We welcome good-faith research and will work with you to verify and fix any issue before it's disclosed publicly.
Before launch
A full security overview will accompany the product at launch. The architecture described here is real and already running through our end-to-end test suite.