CAIN ATLAS, explained for beginners

ATLAS decides whether an AI agent's transaction with another company may go ahead: riskier actions need stronger evidence.

What it is

When an AI agent from one company deals with another company's agent (orders, payments, data), the other side needs to know how far to trust it for this transaction. ATLAS sets the bar: a read at low consequence needs basic identity evidence; a transfer at critical consequence needs the strongest level.

Each transaction is checked: action, consequence and the assurance level the agent can prove. Below the bar: DENY with the reasons. An action type ATLAS does not know is denied and requires the highest level (fixed on 2026-11-03, after our own guide test found it had been treated as the lowest).

An everyday comparison

A bank that needs only a card for a coffee but two IDs and a call-back for a large transfer.

How it works, in 4 steps

1. The transaction states its action, consequence and the evidence the agent has.
2. ATLAS looks up the level required.
3. Enough evidence: AUTHORIZE. Not enough, or unknown: DENY.
4. The reasons are listed.

Try it now

Open the four live demo tools → Every button sends a real request to the live product and shows the real answer.

See it on the homepage: CAIN ATLAS section →

Who it is for

Companies whose agents transact with suppliers', customers' or partners' agents.

FIRME AI CORPORATION · CAIN-42 is a Byzantine trust fabric. Numbers marked "our own check" were run by us, not by an independent party. Nothing on these pages is a certification.