ANVE: the private AI office, on the record. See what that means

ANVE

ANVE maker · updated 2026-10-06

A private GPT for each role.
Grounded in approved office knowledge, not the open internet.

Short answer. A private GPT on ANVE is a role-specific assistant, for example for the accounts team, the front desk or a legal associate, grounded in documents the office has approved and limited to what that role may see. It runs on an appliance on the office network, so questions and answers stay inside the building, and every question is logged against a named user. It answers from the office’s own records and says when something is not in them.

01

What makes it private

  • Runs on hardware the office owns, on its own network.
  • Role-based access: an article clerk does not see the partner’s files.
  • Every question and answer is attributed to a named user in the audit trail.
  • No API keys and no per-token bills for the core suite.
02

Status today

Runs today in our current build, on a development machine. No sealed ANVE unit exists yet; our own unit is in design.

03

Questions people ask

What is a private GPT?
A chat assistant that runs on infrastructure you control and answers from your own documents, instead of a public cloud chatbot. See the glossary entry on private AI.
Can several people share one private GPT?
On ANVE, yes: many named users connect through a browser, each with their own role and permissions, on one appliance.
04

What is not real yet

No sealed ANVE unit exists; every image of the appliance on this site is a labelled render. The software runs today on a development machine. No customers yet, no certifications, no delivery date, and the final price is not announced.

Reservations are ₹999 in India or US$12 elsewhere, refundable in full until you confirm an order in writing (terms). Offices that cannot use cloud AI can apply to be one of ten design partners.