How AI touches your data, and what it never does.

No. We never use customer data to train AI models. That covers everything a customer gives us or lets us read: account details, classification queries, uploaded documents, and data read from the customer's own systems.

This commitment is part of our privacy policy.

We use commercial large language models from OpenAI, Anthropic and Google, called through their APIs, to produce classifications, reasoning and document analysis. Each provider is listed on our subprocessor list.

Each model provider is recorded and risk-assessed in our vendor register, alongside our other vendors.

Yes. Every classification carries its GRI reasoning and the CROSS rulings it relied on, so your team can review the result with the trail attached before relying on it.

Our production application and primary database run on Amazon Web Services in the United States (us-west-2, Oregon). Files customers upload are stored in Google Cloud Storage in the European Union (EU multi-region). Our search index runs on Elastic Cloud in the United States (Google Cloud, us-central1). Each provider is listed on our subprocessor list.

Yes. Our primary production database and its backups are encrypted at rest. Our product APIs are served over HTTPS and accept TLS 1.2 or higher, and the primary database requires encrypted connections.