# Duologs

> Duologs is execution authorisation infrastructure for machine actions — the control layer between machine decisions and real-world actions.

Intent is cheap. Execution is irreversible. The commit moment is where trust must exist.

Duologs governs the commit moment: the point where a machine decision becomes a real-world action. In a Duologs-controlled flow, no high-impact action executes without valid authorisation. We help enterprises define authority boundaries, require authorisation before consequential actions (payments, infrastructure changes, identity grants, sending external communications, regulatory filings), and prove what was authorised after the fact.

Canonical tagline: Authorisation before execution.

## Pages

- [Home](https://www.duologs.ai/): Overview of Duologs and the commit moment thesis.
- [Early Access](https://www.duologs.ai/early-access): Join the early access programme for teams building, deploying or governing consequential AI systems.
- [AI Action Risk Assessment](https://www.duologs.ai/assessment): Executive assessment of agentic AI authority maturity.
- [Observatory](https://www.duologs.ai/observatory): Evidence library of real-world AI execution-risk signals and a diagnostic entry point for enterprise exposure.
- [The 5 Phases of Agentic AI Governance](https://www.duologs.ai/whitepaper): Executive framework for governing AI actions.
- [40 Steps of AI Governance](https://www.duologs.ai/whitepaper/40-steps-of-ai-governance): Implementation checklist.
- [FAQ](https://www.duologs.ai/faq): Answers on what Duologs is, the commit moment, and how authorisation infrastructure differs from monitoring.

## Optional

- [40 Steps of AI Governance (PDF)](https://www.duologs.ai/whitepapers/40-steps-of-ai-governance.pdf): Implementation checklist download.