GovAI Summit 2026: AI in Action

Copilot Pro Foundations in Government

- What Copilot Pro is and how it supports APS work

- Key differences between Copilot Pro and enterprise Copilot

- APS-specific considerations: data flow, FOI, and compliance guardrails

- Lessons learned from the APS Copilot trial

Copilot Pro in Everyday APS Documents & Communication

- Drafting, rewriting, and summarising briefs and reports in Word

- Outlook efficiencies: meeting prep, reply automation, and managing inbox load

- Teams and PowerPoint: generating recaps, action lists, and slides from meeting notes

- Hands-on prompt lab: fixing weak prompts and tailoring outputs to APS tone

Copilot Pro for Numbers, Trends & Reporting

- Using Copilot Pro to analyse finance, contracts, and service delivery data

- Spotting trends, preparing reports, and summarising findings

- Data privacy risks: when not to use Copilot Pro and why

- Guided scenario exercise: solve a reporting challenge with Copilot Pro

Ethical, Safe, and Sustainable Copilot Pro Use

- Managing APS values, FOI, privacy, and bias risks

- Human-in-the-loop practices: when to edit, override, and validate outputs

- Building a role-specific prompt library and reusable templates

- Guided scenario exercise: complete an APS-specific use case end-to-end with Copilot Pro, then team share-back

Join Liquid Learning to launch two days of cutting-edge insights on artificial intelligence in government. Meet peers from across jurisdictions, explore the challenges and opportunities of responsible AI adoption, and get ready for a program focused on practical action and public-value impact.

Discover how the NSW Department of Education conceived, built, and deployed EduChat, an AI-powered chatbot designed to support both students and teachers across NSW. Michelle and Dan will share the story behind this landmark project, from its educational principles and governance to the real-world challenges of embedding AI responsibly in classrooms. Learn how they tackled collaboration between technologists and educators, the guardrails put in place to ensure safe and equitable use, and the lessons learnt about scaling innovation without losing sight of what makes classroom learning so important.

- Understand how clear problem-definition and stakeholder input shaped EduChat, and how to apply this to AI projects across government

- Learn practical ways to balance speed, governance, and ethical guardrails when deploying AI in the public sector

- Take away tips for scaling adoption and measuring impact so AI tools deliver real value for diverse users

Drawing on her experience and expertise as a long-standing advocate for open government, Pia explores how DPI can serve as the bridge between AI ethics and real-world implementation. This session will unpack how governments can embed ethical design, interoperability, and transparency into AI systems to build enduring public trust. Through practical examples and lessons from across sectors, Pia will demonstrate how to move from aspirational principles to accountable, human-centred AI delivery.

- Learn how to operationalise AI ethics through data stewardship, adaptive policy design, and trust frameworks

- Discover practical DPI strategies that align governance, service delivery, and algorithmic accountability

- Gain tools to design AI systems that enhance citizen confidence in digital government

Recharge and connect over coffee with colleagues driving digital transformation across the public sector. A chance to compare notes, swap ideas and spark collaborations before the next round of case studies and strategy sessions.

AI in government is no longer just about high-level ethics, it’s about embedding trust, governance, and innovation into everyday delivery. In this session, Daniel will take participants inside the work of the NSW AI Office as it moves beyond theoretical principles to practical implementation. Explore how to establish effective controls across different AI solutions, assess use-case risks, and build organisational capability as Daniel unpacks how government can learn through communities of practice, re-use common processes, and partner across procurement and policy to responsibly reimagine services.

- Explore a risk-based AI framework that embeds safety and ethics in real-world delivery

- Discover practical pathways for responsible adoption of AI in teams

- Re-use processes, procurement partnerships and communities of practice

Drawing on her leadership in international AI standards and governance, Aurélie Jacquet will unpack how governments can move from fragmented Responsible AI frameworks to a single, scalable assurance model that enables both transparency and acceleration. This session will explore:

- How to unify Responsible AI processes into a single, scalable risk-management and transparency framework.

- Latest international best practice in AI governance and transparency: What Australia can adapt now

- Practical tools for embedding transparency, auditability, and assurance in real projects

Take time to refuel and network. Whether you’re exploring AI governance, service delivery, or data ethics, lunchtime conversations often lead to the partnerships and ideas that drive innovation back in the workplace.

AI is transforming how we work, offering government staff the chance to focus on higher-value tasks and strengthen workplace capability. At the same time, it raises important questions around skills gaps, role design, and cultural change. This panel explores how the APS workforce can harness the benefits of AI while navigating the challenges, preparing people to thrive in an AI-enabled future.

- Building AI literacy at scale without overwhelming staff

- Leveraging AI to reduce pressure on the workforce

- Balancing upskilling existing employees with recruiting new talent

Pause for reflection and refreshment. Discuss key takeaways from the morning’s sessions and exchange perspectives with practitioners tackling similar implementation challenges across government.

Generative AI is only as good as the prompts that shape it. In this practical session Felipe demystifies prompt engineering through a hands-on exploration of how decision-makers and IT professionals can design, test, and refine prompts to generate consistent, high-quality results across a range of organisational use cases. Learn how to move from curiosity to control, mastering the art of structured prompting, context setting, and output evaluation to improve productivity, accuracy, and compliance in AI-enabled workflows.

- Understand how prompt structure, tone, and context influence AI model performance and reliability

- Apply practical frameworks for creating reusable prompt templates tailored to business, data, and technical applications

- Experiment with multi-step and role-based prompting techniques to achieve more accurate and auditable AI outputs

Reflect on the day’s insights and lessons from leaders shaping the AI frontier in public service. Consider what resonated most, and what opportunities you’ll take back to your own department or agency.

Continue the conversation in a relaxed setting. Build meaningful connections with policy, technology, and innovation peers over canapés and drinks as you share ideas about responsible and human-centred AI in action.

Start the day with a focus on application. Gather for an interactive planning session to translate yesterday’s insights into tomorrow’s outcomes, identifying where your team, agency, or community can make real progress.

Lieutenant Colonel Adam Hepworth leads AI exploration for the Australian Army, specialising in autonomous systems, human-machine teaming, and AI-enabled decision-making. In this session, he shares how he is turning advanced AI research into practical, high-impact solutions for complex environments, and the lessons his experience offers for government leaders implementing AI responsibly.

- Translating advanced AI research into real-world impact

- Insights on human-machine collaboration and decision-making

- Strategies for leading responsible and innovative AI initiatives

In this session join Kristin, National Manager Forensics (Acting) at the Australian Federal Police, as she explores the role of AI in forensic and investigative work. This session will focus on the opportunities and considerations of adopting AI in sensitive environments, all while maintaining the highest standards of privacy, security, and evidentiary integrity.

- How AI is shaping the future of forensic and investigative practices

- Practical lessons and insights for implementing AI responsibly

- Lessons learned around AI adoption in complex environments

Grab a coffee, connect with new contacts, and compare notes on the evolving discussions around AI capability, governance and ethical frameworks.

As AI moves from concept to mission-critical implementation, organisations face the challenge of translating responsible AI governance principles into tangible engineering practice. In this session, Liming Zhu explores how CSIRO Data61 is bridging the gap between policy-level governance and the technical foundations of trustworthy AI, showing what it takes to embed transparency, accountability and resilience into AI systems at scale.

- How to operationalise responsible AI principles through data and software engineering design

- Frameworks that connect governance, ethics and assurance with measurable engineering outcomes

- Insights from CSIRO’s data platform revolution on scaling trustworthy AI across complex systems

Dr Leila Alem draws on her 25+ years of experience working at the intersection of academic research, industry and start-up ventures to help founders understand how AI can act as an accelerator of innovation, not simply as a product in itself. In this session, she reframes AI’s role in public-sector programs, showing how it can validate ideas faster, accelerate innovation in mission-driven areas such as climate action, and bridge the gap between research, commercialisation, and real-world implementation.

- Understand how government agencies can integrate AI as a strategic accelerator to test, learn and scale new ideas safely and efficiently

- Explore frameworks that reduce time-to-value in complex digital transformations

- Case studies where AI has augmented human capability or process rather than replacing it

Network with fellow decision-makers and innovators. Take stock of what’s been learned so far and prepare for the final afternoon of forward-looking sessions and real-world use cases.

Governments worldwide are moving quickly to adopt AI, but safe and responsible use requires more than principles. It demands practical governance frameworks and effective risk management. This panel brings together leaders who are translating ethical commitments into real-world practice, ensuring AI adoption in government is safe and trusted.

- Embedding ethics into daily decision-making

- Building agile governance frameworks that keep pace with AI innovation

- Navigating regulatory and compliance requirements

As artificial intelligence transforms the cyber landscape, human judgement and culture remain the most powerful lines of defence. In this session Jo explores how trust, ethics, and diversity underpin cyber-resilient organisations, and why technology alone can’t protect us. Drawing on real-world case studies, she’ll demonstrate how leaders can cultivate awareness, accountability, and adaptive behaviours across their teams to outpace evolving AI-enabled threats.

- Understand how AI is reshaping both the threat landscape and the human factors that influence security outcomes

- Identify the cultural and behavioural levers that strengthen organisational cyber resilience

- Apply practical strategies for embedding awareness, responsibility, and ethical decision-making into everyday operations

Final recharge before closing sessions. A perfect moment to consolidate learnings, share reflections, and identify the collaborations you’ll take forward beyond the summit.

Bring it all together. Summarise insights from two days of knowledge exchange and practical case studies. Walk away with clear actions, renewed energy, and a network of peers committed to building an AI-enabled, citizen-centred public sector.

Understanding AI-Driven Cyber Risk

- Identify how AI systems create new risk surfaces across data pipelines, model training, and algorithmic decision-making

- Recognise emerging threat types including data poisoning, model inversion, and adversarial attacks

- Understand regulatory and ethical implications of insecure AI systems

- Analyse case studies of AI-related cyber incidents and lessons learned

People, Culture and Ethics: Building Human Resilience

- Examine how organisational culture and diversity influence cybersecurity behaviours in AI projects

- Understand the ethical dimensions of AI security, including bias, transparency, and accountability

- Explore practical methods to embed security awareness and psychological safety across teams

- Learn how leadership communication and inclusive practices can strengthen cyber resilience culture

Breaking Silos: Collaboration Across Functions

- Map the key roles and responsibilities across technical, executive, and regulatory stakeholders

- Develop shared language and communication strategies between cyber, data, and governance teams

- Understand how cross-functional risk ownership improves detection, response, and recovery

- Work through examples of effective governance models for AI and cybersecurity integration

From Insight to Action: Designing Resilient AI Systems

- Apply workshop learnings to develop a draft cyber resilience strategy for an AI initiative

- Conduct a scenario-based exercise assessing risk, resilience, and mitigation pathways

- Identify key performance indicators and governance checkpoints for ongoing maturity

- Commit to next steps for embedding resilient design and ethical practice in AI programs