FAQ

Answers, clearly.

Private AI raises practical questions around data, deployment, and control. Here are straightforward answers.

Do you train models on our data?+

No. Your stack is dedicated to your organization. No training on your data, no shared endpoints.

What’s the difference between On-Prem and Secure GPU?+

On-Prem is installed inside your environment for maximum sovereignty. Secure GPU is a dedicated, locked-down instance for teams that want strong privacy without maintaining hardware.

Can the system use our internal documents?+

Yes. We use a local vector database and retrieval pipelines so responses are grounded in your private materials.

What kinds of organizations are a fit?+

Teams where confidentiality is non-negotiable—legal, medical, financial, and any organization handling sensitive records.

Can you build additional agents and workflows?+

Yes. We build purpose-built agents for drafting, intake, summarization, routing, and other scoped workflows.

Still have questions?

No hype. No generic demos. We’ll discuss your systems and your risk profile directly.

Contact IronGate →