AI on the Frontline: Addressing Governance, Ethics, and Safety in the Public Sector
The AI on the Frontline Conference 2025 offered an opportunity to explore the use and deployment of AI tools across the UK public sector and the growing need for robust governance and ethical oversight. The conference emphasised that driving productivity gains is only meaningful when AI systems are deployed safely, ethically, and with a clear commitment to serving societal needs. Bringing together practitioners, policymakers and researchers, the event explored how public sector organisations are navigating these intersecting demands.
Titled “AI on the Frontline: Addressing Governance, Ethics, and Safety in the Public Sector”, the event discussed the need for robust governance and actionable frameworks capable of keeping pace with technological change. Participants shared experiences from local government and wider public sector contexts, illustrating how organisations are navigating questions of accountability, transparency, and responsible use. Beyond discussing the technical deployment of AI, they reflected on the practical challenges encountered in its adoption and the lessons learned in this process.
Across the sessions, the conference examined the evolving landscape of AI governance in the UK through both practical and research-informed perspectives. Presentations considered how local communities make sense of AI-based technologies in everyday environments, how AI ethics is translated into duties and actionable governance processes within English local government, and how AI safety concerns and challenges manifest in the UK public sector. An interactive workshop allowed participants to engage with a practical scenario on social care designed to provoke critical thinking and discussion about the challenges of applying AI responsibly in real-world settings. These discussions underscored the importance of thorough planning, clear guidance, organisational capacity, and continuous monitoring and learning in supporting safe implementation.
Looking ahead, the conference concluded by mapping the “known unknowns” that continue to guide the future of AI in public services, ranging from data stewardship and regulatory alignment to evolving expectations from citizens and communities. Reflections emphasised that while AI holds potential to improve public-sector outcomes, its long-term success depends on sustained commitment to ethical practice and governance. The event closed with a reminder that those on the frontline should use their domain knowledge to shape how AI is implemented for the benefit of people and society, who are at the centre of public services.
The event took place on Wednesday, 19 November 2025, at West Hub, University of Cambridge.
Organisers (arranged alphabetically by surname): Viviana Bastidas, Maya Indira Ganesh, Kristina Khutsishvili, Kwadwo Oti-Sarpong, Jennifer Schooling, and Ruofei Wang.