AI Governance

Omnox uses AI in two places: a small demo on this marketing site (“Brand Brain”), and a set of real product features inside the Omnox CRM (deal coaching, document extraction, email drafting, an AI command bar, and others). Both run on Anthropic's Claude models. This page explains how, so you don't have to take it on faith.

What gets sent to the AI provider

• On this site: only the fixed, preset question text you click in the Brand Brain demo. There's no free-text input, and nothing else you type anywhere on the site is sent to Anthropic.
• Inside the CRM: the specific record content relevant to the feature you invoke at the moment you invoke it, for example, a deal's notes and stage history when you ask for coaching, or the contents of a document you upload for extraction. Nothing is sent in the background or without a triggering action you took.

What doesn't happen

• We don't use your data to train models, ours or anyone else's.
• One customer's CRM data is never used to answer, suggest, or improve anything for another customer. Every AI call inside the CRM is scoped to the tenant that triggered it, the same isolation that applies to the rest of the product.
• AI-generated content isn't published, sent, or acted on automatically. A human reviews and approves AI drafts and suggestions before they go anywhere (an email draft, a deal stage change, a task) unless you've explicitly configured otherwise.

Reliability and limits

AI outputs are informational and productivity aids, not authoritative answers. They can be wrong, incomplete, or confidently stated and still incorrect; this is true of every large language model, not a caveat unique to Omnox. Nothing generated by Omnox's AI features constitutes legal, financial, medical, HR, or other professional advice, and you're responsible for verifying anything you rely on for a real business decision. This is stated more formally in our CRM Terms of Service.

Built with AI assistance

In the spirit of the same honesty: a substantial part of Omnox's own codebase (this site and the CRM) was written with the help of AI coding tools, under human direction and review. We disclose this because we think a vendor's own use of AI in building their product is exactly the kind of thing a careful buyer should be told, not something to obscure.

Your control over AI features

CRM AI features can run on Omnox's shared API key or your own (bring-your-own-key), configurable in Settings. Using your own key means those calls run under your own Anthropic account and its terms, not ours.

Questions

If you need more detail than this page gives you (for a security review, a DPA, or your own AI-governance checklist), email hello@omnox.ai and we'll answer directly rather than pointing you back at this page.