Private AI on secure GPUs
Compute in the dark. Proof in the open.
Run AI on sensitive data, like patient records or legal files. Your data stays encrypted the whole time. Before you send anything, your browser checks a signed report from the hardware, so you don't have to trust us.
How it works
Your browser checks. Not us.
The check runs on your side, so we can't fake it. Our servers only pass messages along.
We start your endpoint
It runs our public software. Anyone can check its fingerprint.
The same software for every customer.
The chips sign a report
The CPU and GPU sign a report of exactly what is running. Only real hardware can sign it.
Signed with Intel, AMD and NVIDIA keys, not ours.
Your browser checks it
Your browser checks the report against our public software. If anything is off, it stops.
Runs on your computer.
Then your key is sent
Your browser locks your key so that only this endpoint can open it.
We pass it along, but we can't read it.
What it proves
What you get, and what you don't.
Clear answers for your security team.
A passed check proves
- Your endpoint runs on real, secure hardware.
- It runs our public software, unchanged.
- The report is new and made for your check.
- Only that endpoint can open your key.
It doesn't prove
- That our software has no bugs. It's small and public, so anyone can review it.
- That we can't switch it off. We still run the machines.
- Protection from every hardware attack. Some advanced attacks are out of scope.