Reimagining web publishing with AI and UX research: Experimenting at the University of Edinburgh

When AI can be applied to anything, where do you start? And how do you learn about AI’s potential in an ethical and responsible way? At the University of Edinburgh, we wanted to learn if AI could help us with some of our website challenges and we turned to UX research to help. With a wealth of information, used by everyone from anticipatory school-leavers to world-renowned academics, our website needs to offer outstanding digital experiences to all, but how do we present the right information, in the right way to everyone? and how can we ensure hundreds of content editors publish web content consistently? In this talk, I’ll explain how we teamed up with open-source Drupal AI experts to explore how AI-powered tools could support smarter, more sustainable content management and web publishing. I’ll cover our UX-research-led approach, and walk through details of a series of controlled experiments we ran with staff and students to understand the possibilities for applying AI and to appreciate the associated risks and implications. You’ll learn what we found – which was both surprising and exciting. AI agents that teach content editors how to use publishing tools in real-time? Prompt engineering to embed style guide rules into content workflows? AI-generated embeddings to enable more intuitive search results? All of these ideas are showing real promise and there’s more to come. For the experimentally curious, including skeptics and enthusiasts alike, I’ll share our iterative experimental framework, as well as a toolkit to design your own AI experiments, drawing on areas like conversational design, learning design, interaction design, service design and AI configuration. You’ll learn how to gather robust data to help you make responsible, considered AI decisions whatever digital experience you’re focused on.