Is my data used to train the AI?

Your questions and data are sent to Claude through Anthropic’s commercial API only to answer your question. Under Anthropic’s commercial terms, API data is not used to train its models by default.

Can the analyst make changes?

No. Its tools are read-only. It can explain and recommend, never change settings, close detections or contact anyone.

Yes. An admin can switch the AI analyst off for the whole organization in Settings.

A security analyst you can ask anything

Ask questions in plain English or Dutch and get answers grounded in your own detections, assets, vulnerabilities and monitors, with the evidence shown.

AI security analyst

Small IT teams don’t lack data, they lack time and a second opinion. Generic AI chatbots don’t know your systems, and pasting logs into them is a privacy risk.

Runs on Claude, with read-only tools for the security summary, threat search, asset risk, incidents, vulnerabilities, monitors and IP reputation.
Shows which tools and data it used for each answer, so you can check the evidence.
Scoped to your organization: it cannot see another customer’s data, and it can explain and recommend but never change anything.
Writes full assessments with owners and timeframes, and the summaries in your reports.
Admins can switch it off for the whole organization; it is also usable from Claude Desktop through the OZAG Security MCP connector.

Is my data used to train the AI?: Your questions and data are sent to Claude through Anthropic’s commercial API only to answer your question. Under Anthropic’s commercial terms, API data is not used to train its models by default.
Can the analyst make changes?: No. Its tools are read-only. It can explain and recommend, never change settings, close detections or contact anyone.
Can I turn it off?: Yes. An admin can switch the AI analyst off for the whole organization in Settings.

OZAG Technologies · KVK 42167370 · Laan van Oversteen 20, 2289 CX Rijswijk · +31 6 24496620 · support@ozagsecurity.com