The OECD.AI Policy Navigator

Our policy navigator is a living repository from more than 80 jurisdictions and organisations. Use the filters to browse initiatives and find what you are looking for.

DCCEEW AI Ethics and Governance Framework


-
Added by:   OECD analyst
Added on:   18 Nov 2026
Updated by:   OECD analyst
Updated on:   18 Nov 2026

The DCCEEW AI Ethics and Governance Framework sets practical rules for the safe, ethical and transparent use of AI, translating whole-of-government policy into departmental requirements. It applies to all staff and contractors, supporting teams with clear safeguards and executives with defined accountability. It was developed to manage AI-specific risks while enabling responsible innovation and public trust.

Initiative overview

The Department of Climate Change, Energy, the Environment and Water (DCCEEW) initiative addresses a governance and risk gap arising from the increasing use of AI within the department, where existing enterprise risk, ICT and policy frameworks do not fully account for AI-specific risks. These include ethical risks (such as bias and unfair impacts), privacy and security risks, lack of transparency or explainability, inappropriate automation of decisions, and potential harm to public trust if AI is poorly governed. At the same time, the initiative seeks to leverage the opportunity for responsible AI adoption, recognising the benefits AI can deliver for efficiency, analysis, decision support and innovation, as well as the risk of not adopting AI where appropriate.

The objectives of the framework are to: 

  • ensure AI is used in a safe, ethical, transparent and accountable manner across DCCEEW;
  • operationalise whole-of-government AI policy and assurance requirements within departmental practice;
  • embed risk-informed decision-making across the AI lifecycle through mandatory assessments, documentation and controls; 
  • clarify roles, responsibilities and accountability for AI use, including escalation of high-risk cases; 
  • and support public trust through transparency and disclosure of AI use.

The framework is intended to be institutionalised through mandatory application to relevant AI use cases, integration with existing enterprise risk and governance arrangements, use of the AI Use Case Register, staff training, and oversight by established governance bodies. It will be reviewed at least annually and updated as technology, risks and whole-of-government policy evolve. Over time, it is expected to mature through the development of supporting guidelines for specific technologies and use cases, increased organisational capability, and closer alignment with evolving national AI assurance approaches. While departmental in scope, the framework is designed so it can be adapted by portfolio agencies, supporting broader consistency in public sector AI governance.

Results, outcomes and impacts

Increased engagement by staff with the document and process. More line areas have registered AI tools they are using or planning to use in the AI Use Case Register. As a result, more projects have been risk-assessed, with identified mitigation activities.

Challenges and lessons learned

Most line areas are asking for the guide to consider the environmental impact of GenAI.

About the policy initiative


Category:

  • Regulations, guidelines and standards

Initiative type:

  • Principles/guidelines/frameworks for trustworthy AI

Status:

  • Active

Start Year:

  • 2025

Binding:

  • Non-binding

Other relevant urls:

—