Product · Messaging
No new app to adopt. The assistant answers inside the platforms your team lives in, the messengers on their phones, your website, your own software. The same identity, permissions and numbered citations follow each person through every doorway.
Whatever the doorway, the conduct is the same — recognise the person, answer inside their permissions, cite the sources, log the exchange. A voice note gets a polite "please type your question", never a guessed answer.
Answer with the member's own access, answer as guest, or stay silent. You choose per group; nothing is inferred.
A sender the deployment cannot match sees only the guest groups you explicitly opened — none by default, on every channel alike.
Each platform's own signature scheme is verified before any processing; unsigned calls are rejected.
They connect outward-only, so nothing of yours is reachable from the internet.
Assistant
Every channel ships with every deployment: the workplace platforms your team lives in, the messengers on their phones, email, the web console, an embeddable website widget and a plain HTTP API for your own software. Each one is optional and off until you turn it on.
No, that is the point. The assistant lives where your team already types. Nobody adopts a new tool; they message the assistant the way they message a colleague.
Colleagues are recognised from their first message: work accounts match automatically, phones link with a pre-registered number or a one-time code. From then on the same identity, permissions and citations follow the person everywhere. The exact mechanics per platform are in the docs.
Nothing, until you decide otherwise. A sender the deployment cannot match is a guest, and guests see only the groups you explicitly opened to them, which are none by default, on every channel.
Group chats follow an explicit policy you choose: answer with the member's own access, answer as guest, or stay silent. Nothing is inferred, and every exchange lands on the same audit record.
Yes. Nothing gets processed unless the platform itself proves the message really came from it, and several channels connect outward-only, exposing nothing of yours. The security page holds the full record if your IT lead wants it.
Every paste into a consumer chatbot is a question your own systems could not answer fast enough. Security vendors sell detection, which reads the app and never the question. What the 2026 breach data actually shows, how far Microsoft's new Shadow AI controls really reach, and why the only fix that scales is making the company answerable.
Glean is Google for your company: it finds where things are stored. Naxis is the private knowledge engine that knows what is actually true right now. Different machines, different questions, different prices. An honest way to choose, and the Glean alternative for companies of 5 to 200 people.
Lidl wrote off €500 million trying. Birmingham City Council is £144 million in and reinstalling from scratch. Consolidation megaprojects fail for a structural reason: real work is too fluid to hold one shape. The part you actually wanted, one place to ask, never needed the migration.
