Digital Ethics Summit 2025

Afternoon Coffee Break

Digital transformation isn't just a technical challenge to be governed or a business tool to be optimised. AI is a once-in-a-generation transformational force reshaping how we work, learn, think, create, connect, and even how understand what it means to be human. Opening the afternoon, this keynote will discuss the tools for responsible AI before exploring what progress and change 2025 ushered in for human-centred AI innovation. It will consider whether we see enough recognition, momentum and focus on building technologies that genuinely serve human flourishing and address real world problems. If not, our keynote will ask: is a rebalance needed, and who and where should take the lead? This fireside conversation will probe the future our current choices are building, how to ensure AI amplifies rather than replaces human agency, and crucially how to keep human values at the core of AI development, deployment, and decision-making. Expect a conversation that challenges attendees to think beyond technical specifications and tools to consider the broader social impact and still-unanswered questions of AI development and adoption.

How can we ensure our AI systems are fair, transparent, and compliant with evolving regulation? How do we move from AI ethics principles to real-world impact? This panel presents the outcomes of the Fairness Innovation Challenge (FIC), a cross-sector initiative delivered in partnership by DSIT, Innovate UK, the ICO, and the EHRC to drive innovative solutions to bias and discrimination in AI systems.

Hear from the four winning projects across higher education, finance, healthcare, and recruitment, as they share how they developed and tested innovative tools to make AI fairer, more transparent, and aligned with emerging regulation. From auditing CV screening algorithms and improving fairness in educational tools, to reducing bias in clinical AI systems and designing LLM fairness toolkits for finance, each project highlights concrete approaches to ethical AI. Through discussion of lessons learned, socio-technical design choices, and regulatory challenges, this session offers a look at what it really takes to assure fairness in AI.

Meet the Responsible AI Practitioners: a view from the frontlines of AI adoption

From collaborative robots on factory floors to full body preventative diagnostics tools like AI mole mapping, we're witnessing the emergence of embodied AI that inhabits and interacts with our physical world. Embodied AI introduces entirely new ethical considerations for workers in sectors and industries across the UK. Human-Computer interactions can now include manufacturing workers adapting to AI-powered robotic systems, healthcare professionals collaborating with diagnostic AI in immersive environments, and service sectors deploying AI that customers experience as embodied beings, rather than software tools. Each interaction demands nuanced, contextual approaches to Human-Computer Interaction that preserve human dignity and agency. This panel will examine how we maintain human-centred design principles when the computer breaks free from the screen and enters the physical space around us. What new forms of transparency and accountability do embodied AI systems require? As AI gains physical form, the importance of ensuring that Human-Computer Interaction works for people has never been greater.

This year, the global AI governance debate has seen an increased focus on AI security. Concerns around AI security are believed to have been a determining factor at the conclusion of the France AI Action Summit. We have also seen the UK's AI Safety Institute renamed the AI Security Institute and the US's AI Safety Institute was renamed the Center for AI Standards and Innovation. This raises the question of whether we have witnessed a fundamental shift in the AI discussion or whether this is simply a matter of semantics.

Bringing together perspectives from leading research institutions, legal practice, policy think tanks, and international governance, this session will unpack the events of the year and consider whether a fundamental shift has occurred, explore what may have been gained or lost and what this means for the digital ethics community moving forward. With the February 2026 AI Impact Summit in India looming the panel will discuss what we might expect on the agenda on issues of safety and security. As geopolitics continue to play a role in the responsible AI discussion, this panel will conclude with a focus on the potential direction of travel for the UK and Europe in this discussion, and to what extent there is a need for countries to come together to offer an alternative approach that balances ethical questions and issues related to both safety and security.

Public sector AI deployment presents significant opportunities for efficiency gains while reshaping how citizens interact with essential services. From NHS diagnostic tools to local government benefit assessments, AI systems in the public sector carry unique responsibilities that require tailored approaches to assurance and ethics. Defence applications particularly requiring careful consideration of autonomous decision-making, accountability, and the ethical implications of AI in security contexts. This session draws on techUK's new sector-specific AI assurance paper to examine pilots and progress in the development and deployment of AI across justice, health, education, and defence. Our panel of frontline practitioners and innovators will share insights on how AI assurance distinguishes successful public sector implementation and how we can tackle barriers such as data silos, legacy systems and procurement processes.

One billion people worldwide experience disability, yet they're often excluded from the datasets and design processes shaping our digital future. How do we change this? What if the next generation of tech could be built differently, designed and developed with disabled people, trained on data that prioritises functional needs over broad categories and generalisations?

Join this interactive breakout session facilitated by the Royal Society exploring how data-driven digital technologies can reduce everyday barriers faced by disabled people – when disabled people are meaningfully involved in the process. The Royal Society's Disability technology report (June 2025) demonstrates that inclusively designed, sustainable digital assistive technologies (whether for work, play, rest, or care), can create a more accessible society for all. In the AI era, built on vast datasets, the report emphasises innovative research methods like "small data" approaches that derive insights from limited data, enabling development of personalised digital assistive technologies.

Attendees will explore key areas including travel, gaming, music, health and social care, and work, examine how current practices support, or fail to support, inclusive data collection, sustainable technology, co-design principles, digital inclusion, and smartphones as assistive devices. On this International Day of Persons with Disabilities this workshop offers opportunities to discuss and share practical approaches to inclusive digital innovation. Participants will benefit from insights that can be applied to a wide range of technology development.

Ministerial Keynote Address

Welcome and Introductory Remarks - Antony Walker, Deputy CEO, techUK

Closing remarks by Antony Walker, Deputy CEO, techUK

This hands-on workshop makes public interest central to AI design from the ground up. Participants will work collaboratively to explore how public interest and participation can support designing and reimagining AI systems to work for people and communities. Through a structured exercise, you'll explore how thinking about public benefits and participatory approaches can and are reshaping AI development.

The session will invite participants to explore what terms like ‘public benefit’ and ‘participation’ mean for them and their organisation, discuss the importance of capturing nuances in what ‘public benefit’ can mean for different publics, get inspired to find ways to translate these visions into their work as we share example projects where visions for public benefit collided in practice, as well as to tips on how to work through different tensions that might arise in the process productively.

Whether you're a technologist seeking more inclusive design methods, a policy maker interested in democratic AI governance, or a community advocate looking to influence AI development, this workshop provides a generative space for collaborative exploration of what ‘public benefit AI’ in practice means, who needs to be heard, and how to make it happen.

End the day off with this closing session which will ask one critical question: does the UK have the right foundations to lead in digital ethics, responsible innovation, AI assurance and ethical foresight at scale going forward?

Given what may arise next year, we will examine three key areas. First, our current institutional capacity, do we have the governance structures, institutional focus, infrastructure, and human expertise needed for this challenge? Second, are we directing resources and funding toward the most impactful initiatives, and if not, what needs to change? Finally, looking further into the future what is our long-term strategy and vision for digital ethics in the UK, where should we be focusing new resources and building new institutions or capability to build robust, sustainable capacity as technology continues to evolve?

With the anniversary of the publication of the UK Government’s AI Opportunities Action Plan just around the corner and our upcoming “Delivering Vision to Value” conference celebrating the anniversary, this panel will also discuss what are the key next steps needed given the commitment to responsible innovation and ask what role the digital ethics community can play in supporting putting the plan into action.

Coffee break

This opening session brings together regulatory, government, industry, and civil society to take stock of 2025’s defining developments in digital ethics and responsible innovation. Our panel of experts will discuss how the digital ethics conversation has evolved, what approaches have worked (and which haven’t), and what lessons may need to shape the year ahead. Making for a very AI-heavy year, the UK government signaled its commitment and investment to building an AI assurance ecosystem – bolstered by the Industrial Strategy and AI Opportunities Action Plan, and industry has leaned into sector-specific assurance. Meanwhile Agentic AI and embodied intelligence have introduced new possibilities and potential risks. This panel will ask and answer whether the momentum is sufficient to propel ethical practices at pace. Or is action still required? Should we have focused this much on AI in 2025 or are there other technologies that also needed our focus? Are we moving in the right direction at the right speed and who must do what, and by when, to close the remaining gaps?

For years, AI ethics discussions focused on the importance of putting principles into practice and being able to prove it. 2025 finally saw that proof come to fruition. Governments, including the UK, EU, and Singapore, along with major global firms and financial institutions integrated AI assurance into their policies and operations. Organisations also report that assurance requirements are increasingly shaping how they approach due diligence for procurement and investment. So, will 2025 be remembered as the year where AI assurance went mainstream? Is it now a distinct professional discipline with real economic impact? In this session, we will draw from techUK’s new Sector Specific AI Assurance report to explore which sectors are leading the way, what this means for the UK AI Assurance market, and how important AI assurance is to driving forward greater trust for AI adoption. We will also explore whether the UK’s leadership to date should be seen as a competitive advantage and whether the UK can position, and maintain, itself as the global leading market for AI assurance.

Lunch

Open and closed AI models each offer distinct advantages for developers. Open-source models can lower barriers to entry and enable broad participation in AI innovation. Closed models can provide controlled environments with built-in safety measures. Right now for AI developers at the cutting edge of AI the choice of whether to use open or closed models isn't binary. The focus is on how to leverage both models as and where appropriate to democratise AI access while managing risks effectively. As a result, most approaches fall somewhere on a spectrum rather than being completely open or completely closed. Given this reality does the discussion, debate and narrative around open and closed models need to change given the important role both approaches currently and will play as AI models continues to evolve? Rather than debating which approach is superior this panel will hear from leading experts that will examine practical strategies for balancing accessibility with responsibility across different development approaches. This session will discuss the levers for government in supporting the spectrum of model approaches, the tradeoffs and safeguards for each, and how each can be used to effectively drive the UK’s advantages in AI innovation.