Artificial Intelligence (AI) Transparency Statement

The Department of Social Services (DSS) is committed to the safe and responsible use of artificial intelligence (AI).

We recognise that AI offers significant opportunities to improve productivity and enhance service delivery for the Australian community.

We govern our use of AI in line with applicable laws and regulations, the Digital Transformation Agency’s (DTA) policy for the responsible use of AI in government (DTA policy), and best practice.

Purpose of AI adoption

The purpose of adopting AI at DSS is to improve how we work and deliver services. AI can help us streamline processes, increase efficiency, and provide better tools for staff to serve the community.

We are committed to using AI in ways that serve the public interest. This means improving service delivery, being transparent, and upholding ethical standards, while keeping safety and trust at the centre of our AI adoption.

Our current AI use cases

Currently, our approved AI use cases are within the:

• Service delivery, Policy and legal and Corporate and enabling domains; and
• Workplace productivity and Analytics for insights usage patterns, of the classification system for AI use(Opens in a new tab/window).

We currently do not use AI within the Decision making and administrative action or Image processing usage patterns, or the Scientific, Compliance and fraud detection, and Law enforcement, intelligence and security domains.

How our staff use AI

Our staff use AI systems predominantly for:

• Summarising documents, reports and meeting notes into key points
• Preparing newsletters, presentations and web content; rewrite drafts for clarity
• Creating agendas and recording action items
• Generating graphs and dashboards from data; summarising insights
• Extracting information and preparing summaries and redacting information
• Drafting and reviewing policy documents for clarity and compliance
• Summarising contract documents and identifying compliance risks
• Developing training guides and making technical content easy to understand
• Summarising risk registers and preparing risk reports with dashboards
• Generating transcripts of public hearings and sessions e.g Senate Estimates
• Speech to text transcription of telephone calls.

AI systems we use

The current AI systems that we utilise are as follows:

1. GovAI Multi-Model – We use the GovAI platform as a secure space for staff to test AI models and to support staff learning. This helps build capability uplift and supports safe trials.

2. Microsoft Copilot via GovTEAMS - We use Microsoft Copilot in the secure GovTEAMS environment.

3. Microsoft 365 Copilot Chat (DSS Internal) – We use Microsoft 365 Copilot Chat within the DSS controlled environment. 

4. Microsoft 365 Copilot (Pilot) – We have commenced a controlled pilot of the full licensed version of Microsoft 365 Copilot within the DSS controlled environment for approved staff cohorts. The pilot is being implemented using a staged enablement approach, with access granted based on risk, privacy, secrecy and business assessments. 

5. Microsoft Copilot Agents and Copilot Studio (Pilot) – We have commenced controlled pilots of Microsoft Copilot Agents within the DSS controlled environment, including Microsoft-provided agents, DSS-developed agents and Copilot Studio developed agents. These capabilities are being evaluated for approved productivity, knowledge management and information analysis use cases and are subject to additional governance and approval requirements.

6. Otter AI – We use Otter AI for a select group of staff, an AI-powered transcription tool designed to convert speech into text.

Consistent with our risk-based approach, access to certain AI capabilities, including the full licensed version of Microsoft 365 Copilot and Copilot Studio developed agents, may be restricted to approved cohorts or use cases following legal, privacy, secrecy and risk assessments.

DSS AI usage does not include use cases that involve direct interaction with the public or that would significantly affect individuals. All outputs from AI tools used internally are reviewed by a human before being applied or acted on. This ensures technology supports DSS operations without replacing human judgment in areas that affect people’s lives. 

Our AI Governance Approach

DSS takes a risk-based approach to the use of AI, focusing on identifying, assessing and managing the risks associated with the adoption and use of AI systems. We have appointed our Chief Information Officer (CIO) as our Accountable Official. We have also appointed a Chief AI Officer (CAIO) to provide strategic leadership and oversight.

We have established a comprehensive AI governance framework that ensures AI is used safely, ethically, transparently and accountably across DSS. Under this framework, AI is governed through a use case-based model supported by dedicated governance committees, risk and assurance processes, mandatory legal, privacy, security and impact assessments, ongoing monitoring, and clear accountability arrangements. 

Every proposed AI use case must undergo a structured risk, legal, privacy and security assessment and receive approval through the department’s governance processes before it can be procured, developed, deployed or used. AI systems are not approved in isolation; approval applies only to specific, defined use cases, and use outside an approved use case is prohibited. Only approved AI use cases may be used within DSS, with each approved use case remaining subject to ongoing assurance, oversight and review throughout its lifecycle.

As part of this framework, we have an AI Governance Committee to oversee the adoption and use of AI within DSS. The Committee’s functions include:

• guiding the department’s AI strategy, principles and policies
• identifying, assessing and managing AI risks and opportunities
• reviewing and approving AI use cases
• monitoring the performance and impact of AI solutions.

The Committee is chaired by the Chief Operating Officer and includes senior representatives from legal, data, corporate, communications and ICT.

Additionally, a range of existing DSS policies apply to the use of AI. These include policies on privacy, confidentiality, risk management, procurement, archiving, acceptable use of technology, and cyber security. 

This approach ensures that our AI use complies with legal and ethical obligations, protects sensitive information, manages risks effectively, and meets requirements for record-keeping and accountability.

Once implemented, we monitor the effectiveness of AI systems through robust governance arrangements, policies and processes, and by tracking AI usage. We also engage with staff to understand the impacts on workflows and productivity.

Our compliance with the DTA policy

The following table sets out an overview of our compliance with the mandatory requirements of the DTA policy and APS AI Plan:

This statement is authorised by DSS’s accountable official, the Chief Information Officer.

This AI Transparency Statement was updated on 22 July 2026.

More information

• Visit the Digital Transformation Agency to read the DTA policy(Opens in a new tab/window).
• If you have questions or feedback, email AI@dss.gov.au.