CognaOne makes governance the fast lane to production.

One platform for AI systems, agents, risks and controls, with evidence ready for regulators.

See it governHow it connects

How it works

Start where your AI programme is today: see what you run, prove it is controlled, then scale and keep pace.

  1. 01Start

    Connect platforms read-only, register what teams declare and give every AI system an owner.

  2. 02Prove

    Tier the risk against your CPS 220 risk appetite, map controls to CPS 230, CPS 234 and FAR, and collect the evidence.

  3. 03Scale

    Bring agents into production with defined rights and every decision on record.

  4. 04Evolve

    Map new obligations, reassess systems when they change and keep the evidence current.

Built to connect

CognaOne sits between your AI systems and the platforms they reach, giving every team one consistent framework of governance, risk, controls and assurance.

  1. AI systems

    Agents, copilots, models, MCP servers and vendor AI, declared and discovered.

  2. CognaOne

    Governance, risk, controls and assurance on one record, between your AI and everything it reaches.

  3. Platforms and data

    AWS, Azure, Databricks, Snowflake and the rest, connected read-only through connectors and the CognaOne API.

CognaOne API
Platforms and data
AWS BedrockAzure AIDatabricksGoogle Vertex AISnowflakeMicrosoft CopilotMCP serversOpenTelemetryGitHub
CognaOne
GovernanceOwners, policies and approvals on every AI systemRiskSix dimensions and a tier you can explainControlsMapped once, seen on every system it governsAssuranceEvidence that climbs from declared to effective
CognaOne
AI systems
AgentCopilotModelMCP serverEndpointVendor AI

Discover, govern, enforce

Technology, risk and audit teams work from one record, from the first sighting of an AI system to the action it takes.

Ready for APRA's AI expectations (April 2026) and the Privacy Act's automated decision rules, from 10 December 2026.

Every AI system, accounted for

Registered systems sit beside the ones found on your platforms, in traffic and inside vendor products, so shadow AI shows. Each carries an owner from design to retirement.

A risk tier the board can challenge

Six dimensions, a written rule for the tier and an accountable executive on every system, all set against the risk appetite your Board approved under CPS 220. The tier sets which controls apply and how often it is reviewed.

Map once, evidence everywhere

A control mapped once to CPS 230, CPS 234 or FAR (the Financial Accountability Regime) shows on every AI system it governs. 81 controls come ready, so the second line starts from a library, not a blank page.

Assurance that doesn't wait for the audit

Point-in-time testing misses models that drift. Each control climbs from declared to effective only as live evidence supports it.

A person on every high-risk decision

CognaOne checks state, controls and approvals before an agent acts, then allows it, routes it to the named person or blocks it. Every decision replays.

Least privilege for AI agents

Each agent's identities, tools and MCP (Model Context Protocol) permissions sit on the same record as risk and controls. This is where access control catches up with non-human actors.

Let's connect

Tell us where your AI programme is today. We scope the first engagement to your own AI estate, and every enquiry is handled in confidence.

Contact us