This month we joined the Future Facing Learning and AI in Higher Education 2026 conference, hosted by Teesside University, to share what we've learned from our work with the University of Liverpool's Centre for Innovation in Education.

The theme of our session — From Pilot to Pedagogy — captures a shift we think is central to where AI in higher education is heading. It's relatively easy to run an AI experiment. It's much harder, and far more valuable, to move from a standalone experiment to curriculum-aligned support that lives inside the university's own VLE and behaves as an extension of teaching rather than a distraction from it. That's the journey we walked through on stage: what changes when AI is embedded directly in Canvas, aligned to real course materials, and governed with the institution in control.

Just as importantly, we shared what the student data tells you once you make that move. Across the Liverpool pilot, a consistent picture emerged: students reach for curriculum-aligned support at the moments they most need it, often outside timetabled hours and around assessment points, when traditional channels aren't available. That pattern is one of the clearest arguments we've seen for treating AI not as a chatbot bolted onto the edge of a course, but as a genuine pedagogical partner embedded within it.

Alongside the session, our poster — Beyond Chatbots: Curriculum-Aligned AI as a Pedagogical Partner — set out the benefits we're seeing for students, staff, and institutions alike: immediate, contextual support for learners; better insight into student misconceptions for teaching staff; and a safe, governed route to AI adoption at the institutional level.

Our thanks to Teesside University and the organisers such a well-run conference, and to everyone who visited our stand, joined the session, or stopped to talk with us across the three days.