Trust & security

Answers you can defend. Data that stays yours.

A tool that might quietly invent a requirement is useless in a safety review. Adducia is built so the engineer stays in control and can always check the work.

How it earns trust

Source-backed by construction Every artifact ID and claim in an answer is checked against the underlying graph built from your documents. It will not invent a requirement, a standard, or an artifact number. The source is attached so you can verify it yourself.
Honest about its limits When the documents don't support an answer, Adducia says there is no source-backed answer instead of guessing. A confident wrong answer is the most dangerous thing a safety tool can produce, so it is built not to give one.
Your data stays in your environment Adducia ships as an on-prem build that installs on your own hardware and calls the model with your own key, so the only outbound connection is to your own model endpoint. Nothing is sent to us and there is no telemetry, which means material governed by export-control and handling regimes such as ITAR, EAR, or CUI can stay inside your boundary. Anything shown publicly, including any demo, uses synthetic data that is clearly labeled as such.
Human in the loop Low-confidence reads and anything pulled from a diagram are flagged for review rather than silently merged into the graph. The engineer decides what to trust. Adducia's job is to do the finding, not to replace the judgment.
This site itself is a static page with no backend, no database, and no access to any program data. Formal security attestations will follow as we work with design partners; if you have specific requirements, raise them when you get in touch and we will speak to them directly.

Next step

Have a security requirement in mind?

Tell us how your team needs to handle controlled data, and we'll walk through how Adducia fits.

Get in touch