Programme
Sessions marked with a expand — select one to read its abstract, or the papers being presented.
14 October 2026
Professor Ann Kristin Glenster
Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Professor Andy Neely
Chair of Cambridge University Press & Assessment Board, University of Cambridge
Aanya Niaz
AWS Education Equity Initiative, Amazon Web Services
Christopher Ott
Head of AWS Academy, Amazon Web Services
15 October 2026
Professor Ann Kristin Glenster
Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Megan Ennion
Research Development Officer, Digital Education Futures Initiative (DEFI), Hughes Hall, University of Cambridge
Dr Kevin Martin
Moderator · Digital Education Futures Initiative (DEFI), Hughes Hall
About this keynote
As AI becomes embedded across educational environments, fundamental questions arise that extend far beyond measurable educational outcomes. Learning is not merely the acquisition of answers. It involves the cultivation of attention, understanding, independent thought, intellectual curiosity, creativity, capacity for social relationships, and the ability to evaluate, challenge, and generate knowledge claims.
In their keynote, Professor Ann Kristin Glenster and Megan Ennion introduce the concept of a Learner Epistemic Rights Charter to identify, protect, and strengthen the conditions through which learners become knowers. Developed through ongoing research at LOGOS, the Charter seeks to identify and protect the epistemic conditions necessary for learning, curiosity, discovery, creativity, and human flourishing in an AI-mediated world.
Emerging evidence suggests that many of the conditions that enable the full range of human learning may be altered, reshaped, or undermined by AI systems. As increasingly powerful technologies reshape how knowledge is produced, accessed, shared, and applied, they also reshape the structures through which human learners and teachers create and benefit from future knowledge systems. Protecting and strengthening these conditions therefore becomes crucial to securing and ensuring a flourishing epistemic future.
By reframing AI through the lens of learner epistemic rights, the enquiry shifts from how AI can support existing models of education to what conditions are required for learners to explore, imagine, discover, and contribute to new domains of knowledge. Rather than focusing on what AI can do, the Charter asks what forms of human development educational environments should protect and cultivate in an age of increasingly powerful technologies. It offers a conceptual framework for thinking about how human epistemic agency can continue to flourish as new technological frontiers emerge.
Moderator
Megan Ennion, Research Development Officer, Digital Education Futures Initiative (DEFI), Hughes Hall, University of Cambridge
Presentations
- From cognitive necessity to cognitive choice: a framework for human capability in AI-mediated environments
- But-ChatGPT-Saids and the epistemic rights of professional knowers
- Human judgment as curriculum: epistemic agency and the irreplaceable in graduate analytics education
Moderator
Dr Carla Pastorino, Principal Research Manager IELTS, Cambridge University Press & Assessment
Presentations
- Evaluating the impact of an AI interview platform (AVA) on medical school applicant outcomes: access and equity in a multi-institution randomised controlled trial
- Introducing AI through purpose: lessons from a community-based outreach programme with home-educated neurodivergent learners
- From symbols to AI: teacher and SENDCo perspectives on the future of communication support in inclusive classrooms
- Cognitive orchestration, not just offloading: neurodivergent writers, AI scaffolding, and human agency in think-aloud protocols
Moderator
Dr Emily Black, Director of AI Strategy & Enablement, English, Cambridge University Press & Assessment
Presentations
- From performance to purpose: critical AI literacy and the relational foundations of human intelligence in education
- Learning as ontological emergence: what three studies of AI in education reveal about the relational nature of learning
- Learning first, AI second: extending the Triple E Framework for human-centred education
Moderator
Dr Tony Thompson-Starkey, Head of Data, Analytics & AI, Education Futures, Cambridge University Press & Assessment
Presentations
- Privacy-preserving personalisation: designing a responsible AI study assistant for higher education
- Ethical weight in an AI-mediated classroom: from generation to judgment
- Marking with machines, controlling with humans: responsible AI for scalable assessment
- Implementing responsible AI through assessment reform: investigating staff capability during the rollout of a university wide GenAI assessment framework
Moderator
Sarah Hughes, Head of Research (Futures & Innovation), Education Futures, Cambridge University Press & Assessment
Presentations
- Assessing the thinker, not the text: a transparent AI viva for the age of indistinguishable writing
- When AI can perform the work, what is actually being assessed? Construct validity, human judgment, and assessment redesign in higher education
- Design a map-structured AI agent framework to support faculty agency in assessment design
Moderator
Dr Peter Bannister, Postdoctoral Researcher; LOGOS Affiliate Researcher, Universidad Internacional de La Rioja; University of Cambridge
Presentations
- What is the teacher now for? Sustaining epistemic agency in student–AI interactions
- Cultivating human judgment through AI-mediated geopolitical simulations: agency and decision-making under uncertainty
- Peekaboo me: AI conversation and cognitive sovereignty
Moderator
Alberto Garcia, Murray Edwards College, University of Cambridge
Presentations
- The AI-mediated research process: a new experience for doctoral students
- Zero click research: are AI answer tools changing how students engage with primary literature
- Human-AI co-scaffolding in Hong Kong K-12 ESL education: scaling personalized instruction and sustainable tutorial support
Moderator
Dr Carla Pastorino, Principal Research Manager IELTS, Cambridge University Press & Assessment
Presentations
- Building AI confidence in schools and colleges: early evidence from a structured implementation framework
- Practitioner-led inquiry: a learner’s right to think in an AI-rich education system
- Built to fail some: artificial intelligence and the architecture of educational exclusion
Moderator
Dr Emily Black, Director of AI Strategy & Enablement, English, Cambridge University Press & Assessment
Presentations
- Human oversight theatre: why human presence does not guarantee human judgment in AI-assisted education
- AI-enabled digital twins as responsible, human-centred educational infrastructure for higher education learning spaces
- Teachers, AI, and professional judgement: what AI use and non-use reveal about teaching
Moderator
Dr Yasin Karatay, Senior Research Manager, Cambridge University Press & Assessment
Presentations
- Metaphors matter: how conceptions of generative AI shape assessment decisions in higher education
- Knowing when to trust the mark: confidence calibration for AI-first assessment
- What even is authorship? Generative AI as a reckoning, not a rupture
Jane Mann
Chair · Chief Education Officer, Cambridge University Press & Assessment
Professor Rachel Adams
Executive Director and Research Professor, Leverhulme Centre for the Future of Intelligence, University of Cambridge
Dr Jonnie Penn
Associate Teaching Professor of AI Ethics and Society, University of Cambridge
Ali Shah
Managing Director, Responsible AI, Accenture Ltd, Independent Board Member, Ada Lovelace Institute
About this panel
For more than a century, education systems have been shaped by a powerful social contract: education leads to employment, economic participation and opportunity. But as artificial intelligence reshapes the nature of work and increasingly takes on tasks once considered uniquely human, that relationship is being called into question. If work becomes less central to how individuals create value, meaning and identity, what should education be for?
Chaired by Jane Mann, Chief Education Officer at Cambridge University Press & Assessment, this panel brings together leading voices from technology, research and society to explore one of the most important questions facing education today. Together, the panellists will examine how AI may challenge long-held assumptions about work, human capability, and educational success, and consider what it would mean to place human flourishing, rather than economic productivity alone, at the centre of educational purpose.
Moderator
Jonathan Sansom, Director of Digital Strategy, Hills Road Sixth Form College
Presentations
- The snapshot problem: toward longitudinal measurement of human capability in AI-mediated learning
- Getting beyond ourselves: investigating marker “strategies” with eye tracking and AI
- L2-Bench: an evaluation benchmark for measuring LLM capabilities in second language education
- Adapting human rating scales for English speaking assessments with AI as the interlocutor
Moderator
Dr Peter Bannister, Postdoctoral Researcher; LOGOS Affiliate Researcher, Universidad Internacional de La Rioja; University of Cambridge
Presentations
- Cultivating metacognitive career agency: AI, access and human judgement in career education
- Rethinking critical thinking in the age of AI: from cognitive offloading to cognitive scaffolding
- The learner in the AI-mediated epistemic realm: toward cultivating an agentic stance
Professor Ann Kristin Glenster
Chair · Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Carla Aerts
Advisor AIEd Futures of Education and Learning, LOGOS affiliate Cambridge University CRASSH
Dr Jon-Hans Coetzer
Senior Programme Specialist, United Nations Institute for Training and Research
Anders Rapp
Co-founder and COO, Keenious
About this roundtable
AI technologies cross borders far more easily than education policies do. Tools developed in one country are increasingly adopted in others, carrying assumptions about language, knowledge, pedagogy, assessment and the purposes of education. Yet governments and education systems are responding from very different cultural, economic, regulatory and technological contexts, with unequal capacities to evaluate, govern and benefit from these systems.
This roundtable will consider which aspects of AI and education policy can usefully be shared across borders, and which must remain responsive to national and local contexts.
Sarah Hughes
Chair · Head of Research (Futures & Innovation), Education Futures, Cambridge University Press & Assessment
Simon Armitage
Director of International Education, The Perse School Cambridge
Megan Ennion
Research Development Officer, Digital Education Futures Initiative (DEFI), Hughes Hall, University of Cambridge
Dr Evelina Galaczi
Director of Research, English, Cambridge University Press & Assessment
Kelly Webb-Davies
Lead Education AI Consultant, University of Oxford
About this roundtable
Generative AI has brought the limitations of current assessment models into sharp focus, challenging the assumed relationship between the quality of a submitted product and the knowledge, understanding and reasoning of the learner who produces it. This shifts the purpose of assessment from judging what learners produce to evaluating evidence of their thinking, learning processes and decision-making. At the same time, there is a growing need for more sophisticated approaches to reporting learner achievement; the assessment of a broader range of skills and aptitudes; the evaluation of collective and collaborative performance; and the use of immersive, interactive and personalised forms of assessment.
This roundtable will explore alternative and complementary approaches to assessing learning, addressing questions such as: Could assessment attend more closely to the processes through which learners arrive at an outcome, and what does that evidence tell us about student capabilities? What role might dialogue play, whether through questioning during learning or viva-style assessment? How might we recognise collaborative and collective learning? Should assessment also capture dimensions such as effort, motivation, metacognition and self-regulation? What might be gained by replacing a single grade with a richer learner profile?
Moderator
Alberto Garcia, Murray Edwards College, University of Cambridge
Presentations
- A new voice in the classroom: how AI avatar tutors shape the learner-teacher relationship
- Beyond decision support: AI as a co-creator and reflective partner in entrepreneurial life design
- When GenAI joins the classroom community: a critical review of its impact on conceptual understanding, collaboration, and belonging
- Beyond artificial intelligence: designing AI systems that cultivate human intelligence through reflection and empathy
About this keynote
What we believe intelligence to be shapes what we choose to cultivate, measure and automate. Our notion of intelligence has been degraded by simplistic ideas of artificial intelligence largely driven by commercial interests.
Our fascination with artificial intelligence stems from the perceived uniqueness of human intelligence. Fears of AI concern not only how it invades our digital lives, but the implied threat of an intelligence that displaces us from our position at the centre of the world.
Atomism, proposed by Democritus, suggested it was impossible to continue dividing matter into ever smaller components: eventually we reach a point where a cut cannot be made. In the same way, by slicing away at the facets of human intelligence that can be replaced by machines, AI uncovers what is left: an indivisible core that is the essence of humanity.
This keynote contrasts our own evolved, locked-in, embodied intelligence with the capabilities of machine intelligence. It asks what current AI systems can and cannot tell us about human intelligence; and what that means for how we educate in the age of AI.
Paul Angell
Portfolio Lead, Google DeepMind
Professor Ann Kristin Glenster
Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Peter Phillips
Chief Executive, Cambridge University Press & Assessment
16 October 2026
Dr Jing Xu
Chair · Head of Propositions Research, Cambridge University Press & Assessment
Rachel Arthur
Chief Learning Officer, Raspberry Pi Foundation
Professor Ann Kristin Glenster
Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Dr Kevin Martin
Digital Education Futures Initiative (DEFI), Hughes Hall
About this panel
As AI continues to reshape education, what questions should research and development address next? This panel brings together three Cambridge initiatives for a forward-looking conversation about the emerging challenges, possibilities and priorities likely to shape their work over the coming year.
The discussion will explore where research is most urgently needed, what new forms of development and collaboration could be valuable, and what conditions are required to translate ideas into meaningful educational practice. Looking ahead to the next AI and Education Summit, the panel will consider what progress might look like and where Cambridge’s AI and education community might go next.
The conversation will be opened to the audience through questions, reflections and ideas for the year ahead.
Professor Johanna Pirker
Computer Science Professor, Graz University of Technology and the Technical University of Munich
Dr Imogen Casebourne
Moderator · Digital Education Futures Initiative (DEFI), Hughes Hall, University of Cambridge
About this keynote
Artificial intelligence is rapidly changing how we create, play and learn. Games and virtual environments offer a particularly powerful lens for exploring this transformation: they can provide adaptive, interactive and highly engaging learning experiences, but they also raise important questions about how AI affects creativity, critical thinking, motivation and human agency. This keynote explores the opportunities and risks of AI-enhanced learning through games and virtual environments. It asks not only what AI can do for education, but what we need to understand about how people actually learn with AI and how we can design these technologies to strengthen rather than diminish the skills and experiences that matter most.
Dr Jon-Hans Coetzer
Senior Programme Specialist, United Nations Institute for Training and Research
Professor Ann Kristin Glenster
Founder and PI, LOGOS, University of Cambridge and Executive Director Glenlead Centre
Kellie Mote
Programme Lead (Accessibility), Jisc
About this session
This session contains two fireside chats exploring two frontiers for AI in education and human intelligence.
First, Professor Ann Kristin Glenster and Kellie Mote will discuss how neurotechnology could transform education, starting with the benefits these technologies may have for accessibility to how they might pose serious risks related to mental privacy, surveillance, commercial exploitation, inequality, autonomy, and control. The discussion will consider what educators, researchers, and policymakers should begin preparing for before neurotechnology enters education at scale.
Second, Professor Ann Kristin Glenster and Dr Jon-Hans Coetzer will discuss the new United Nations Institute for Capability and Learning Innovation, and the opportunities this presents for the development and integration of epistemic rights in the UN’s approach to innovation in education.
Moderator
Dr Emily Black, Director of AI Strategy & Enablement, English, Cambridge University Press & Assessment
Presentations
- Educating for the AI precipice: capabilities for navigating AI catastrophic and existential risks
- Sprinting with AI: developing collaborative human AI competencies in education
- Cultivating critical thinking and integrity in the age of artificial intelligence
- Student co-creation of an AI ethics, inclusion & society MOOC
Moderator
Dr Hilla Tal, Research, Impact and Learning, LOGOS
Presentations
- Children's voices on thinking in the age of AI: what government school students in India believe about intelligence, agency, and authorship
- Beyond content routing: architecting process-aware scaffolding for inclusive AI in primary education
- Developing, measuring, and evaluating an AI tutor for computational thinking: evidence from a pre-pilot and cluster-randomised trial in Uruguayan public primary schools
Moderator
Dr Carla Pastorino, Principal Research Manager IELTS, Cambridge University Press & Assessment
Presentations
- Rethinking programming education in the age of AI agents
- Personal AI twins in legal education: responsible implementation, AI literacy, and human-centred pedagogy
- Responsible AI-enabled learning for protecting judgement, agency and foundational learning at scale
- When the tutor becomes the solver: strategic user adaptation and responsible AI implementation in education
Dr Tony Thompson-Starkey
Chair · Head of Data, Analytics & AI, Education Futures, Cambridge University Press & Assessment
Alastair Amos
Secondary Mathematics Teacher and Researcher, Leventhorpe School
Professor Alastair Beresford
Professor of Computer Security and Head of the Department of Computer Science and Technology, University of Cambridge
Holly Giles
Student, Hills Road Sixth Form College
Dr Kim R Sylwander
Postdoctoral Research Officer at the Digital Futures for Children Centre in the Department of Media and Communications, London School of Economics and Political Science (LSE)
About this roundtable
AI has the potential to extend personalised learning to millions of learners worldwide. Yet as educational technologies become more powerful, a fundamental question remains: whose needs and interests are they serving? This roundtable explores how learners, educators and researchers can play a more active role in shaping the technologies that influence teaching and learning. Discussion will cover learner agency, children's rights, educational outcomes, governance and the measures of success that matter most in education.
Amin Neghavati
Moderator · Director of Education Futures, Cambridge University Press & Assessment
Daniel Emmerson
Executive Director, Good Future Foundation
Alison Gellett
The Big AI Project Lead, Oasis Big Education Institute
Professor Rose Luckin
Emeritus Professor of Learner Centred Design, University College London
Jane Mann
Chief Education Officer, Cambridge University Press & Assessment
About this keynote
In her keynote, Dr Kristina Kallas, Minister of Education and Research of Estonia, will explore how education and society must adapt to the rapid advancement of artificial intelligence. As AI increasingly performs tasks based on information retrieval and routine analysis, the question is no longer whether to introduce AI into education, but how education itself should evolve in a world where AI is already embedded in daily life. Critical thinking, creativity, problem-solving, and sound judgement are becoming essential not only for learning and work, but also for democratic participation and societal resilience.
Estonia, known for its highly digitalised society and strong education system, is addressing this challenge through its national AI Leap (TI-Hüpe) programme, launched in 2025. The initiative goes beyond providing access to generative AI. It focuses on strengthening teachers’ capabilities, developing students’ AI literacy, supporting education-focused AI solutions, addressing data protection and language-related challenges, and assessing AI’s impact on learning outcomes. Its broader goal is to ensure that students learn not simply to obtain answers from AI, but to think critically, independently, and responsibly while using it.
Minister Kallas has been a driving force behind AI Leap and wider reforms to Estonia’s education system. Drawing on her academic background in political science, multilingual education, and social integration, she advocates a human-centred approach that keeps teachers and learners at the heart of technological change. Estonia’s experience offers a valuable example of how governments can rethink education, skills, and human capital policy in response to AI, while strengthening national competitiveness, innovation capacity, and economic resilience.
Dr Tony Thompson-Starkey
Head of Data, Analytics & AI, Education Futures, Cambridge University Press & Assessment
Anamika Ved
GenAI Product Lead, Education Futures, Cambridge University Press & Assessment
About this workshop
By 2030, an AI agent will sit between most learners and most of what they might learn: planning pathways, selecting resources, and deciding what appears as an option at all. This working session asks what that layer fails to see, and who decides.
We begin with what is already deployed — not forecasts, but agents running today inside the platforms schools and universities already pay for. You will then test a live Socratic tutor on your own phone, and spend eight minutes trying to break it: it is built to refuse to give answers, and your job is to make it. Finally, in small groups, you will interrogate five scenarios from 2035, work out who has become invisible in each.
Bring a phone. No institutional, confidential or learner data is used at any point.
Dr Imogen Casebourne
Innovation Lab Research Lead at the Digital Education Futures Initiative (DEFI), Hughes Hall, University of Cambridge
Professor (Dr) Bryan Maddox
Research Director for Digital Assessment Futures, DEFI, Hughes Hall, University of Cambridge
About this workshop
There is evidence for the effectiveness of dialogic classroom teaching for a range of subjects, and for inquiry-based learning as a means of fostering 21st century and future skills. However, involving children in in-person group-based learning and discussion brings its own set of challenges for teachers. One challenge is that some children may find it more difficult than others to follow and participate in group talk, for a range of reasons. Additionally, it remains time-consuming for teachers to identify and measure progress when using these approaches.
Novel means of engaging with AI, not always specifically developed for education, might play a future role in supporting inclusive dialogic and collaborative inquiry-based learning, but there remain a series of pressing ethical questions to be considered and addressed.
This interactive workshop will involve hands-on demonstration of and interaction with a range of possible approaches to using AI to support inclusive dialogic and collaborative learning. It will involve participants in ethical discussions about the desirability of this, considering both the pressing ethical questions that should shape immediate practice and the potential longer-term implications, both intended and unintended.
This workshop will draw on an ethical framework intended to inform the design and deployment of novel technologies (Casebourne, Mouta, Zhongling, In Press) which calls on designers and educators to consider longer term and possibly unintended consequences of the use of novel technology in education. It also draws on work on a policy paper (underway) about approaches to the implementation of wearable AI to support inclusive collaborative work, as well as on a series of studies into the plausibility of using AI to code classroom talk.
Dr Ingunn Johanne Ness
Senior Researcher and Theme Leader, SLATE, University of Bergen
Dr André Mestre
Research Fellow, SLATE, University of Bergen
About this workshop
Creative Learning and AI: Building a Shared Manifesto invites participants to consider the place and role of generative AI in the creative learning classroom. Drawing on a model that describes creative learning as agentic, affective, relational, and practical, the workshop will support participants in developing principles for the creative and ethical use of AI that promote human flourishing. Participants will explore how generative AI can support experimentation, surprise, and new forms of meaning-making in higher education. Through interactive exercises, they will engage with practical dilemmas, negotiate competing educational values, and identify responsible and pedagogically meaningful approaches to AI use. The session will culminate in a shared manifesto articulating what educators should protect, cultivate, resist, and make visible when designing creative learning with AI.
Ben Hall
Senior Learning Manager, Raspberry Pi Foundation
Alex Parry
Learning Manager, Raspberry Pi Foundation
Sonja Povey
Community Engagement Coordinator, Raspberry Pi Foundation
About this workshop
Join the Raspberry Pi Foundation for an interactive workshop exploring our free AI learning resources. In this hands-on session, you will step into the shoes of a learner, testing engaging activities designed to inspire young people to create with technology.
Moderator
Dr Yasin Karatay, Senior Research Manager, Cambridge University Press & Assessment
Presentations
- LLMs for feedback on L2 writing
- Rethinking assessment paradigms in response to increasingly capable artificial intelligence
- Beyond the essay: designathons as assessment for an AI-enabled world
- Beyond AI detection: designing evidence of authentic learning in online higher education
About this keynote
For generations, we have told young people that education matters because it leads to qualifications, jobs and income. AI is putting that promise under strain. It makes the future usefulness of what we teach harder to predict, and the products of learning easier to obtain without the learning itself. This closing keynote argues that the answer is not to chase whatever machines cannot yet do. It is to take human flourishing seriously as education’s purpose, and to ask what follows for teaching, assessment and the place of AI in the classroom.