2.2 AI for Defence · Module 2 · The Australian AI Policy Stack

Defence's Own Rules: Obligations & the DAIC

The three obligations, who's accountable, and how Defence actually uses AI today.

Defence policy~15 min

The big idea

The Policy Settings for Responsible Use of AI in Defence (March 2026) set three obligations: lawfulness, adherence to values-based principles (accountability, bias/harm mitigation, explainability, reliability and security, human-centric), and proportionate, risk-based controls. It defines the Accountable Officer — designated personnel who keep accountability across the AI lifecycle even when authority is delegated. Governance runs through the Defence AI Centre (DAIC), established July 2024. And Defence is now a real operational user: Microsoft 365 Copilot was rolled out to tens of thousands of staff on its PROTECTED network from September 2025.

Key insight: Delegating a task never delegates accountability. Under the Defence Policy Settings, the Accountable Officer remains responsible for AI use through its whole lifecycle.

What you'll see

Defence's three obligations and the role of the Defence AI Centre.

What the Defence settings establish

The obligations, the roles and the reality. Tap to expand.

Key points to remember

Three obligations: lawfulness, values-based principles, proportionate controls.
Accountability stays with the Accountable Officer across the lifecycle.
The DAIC (since July 2024) is Defence's AI governance hub.
Defence runs Copilot on its PROTECTED network (since Sept 2025).