When students at Queen Mary University of London sit down to learn, they are increasingly sharing the classroom with an unseen collaborator: artificial intelligence. From lecture halls to online platforms, the university is quietly rewiring the learning experience, testing how AI can personalise education, support overworked staff, and reshape the customary rhythms of campus life.
As universities around the world grapple with the implications of generative AI, Queen Mary has moved early and deliberately. Researchers and lecturers across disciplines are embedding bright tools into teaching, assessment and student support, while also confronting thorny questions about ethics, equity and academic integrity. The institution has become a testbed for what higher education might look like when algorithms help design courses, mark assignments and guide students through complex material.
This is not a story of robots replacing lecturers, but of a research-intensive university trying to harness AI’s promise without losing sight of its risks.In labs, classrooms and policy meetings, Queen Mary is experimenting with new models of learning that could set a template for the sector-if it can strike the right balance between innovation and obligation.
Harnessing artificial intelligence to personalise learning experiences at Queen Mary University of London
In laboratories, lecture theatres and virtual classrooms across campus, data from students’ learning journeys is being transformed into tailored academic support. Predictive algorithms analyse patterns in quiz performance, attendance and engagement on the virtual learning surroundings to surface timely recommendations, helping students revise core concepts before assessments and discover extension materials when they excel. These systems are designed to be assistive rather than intrusive, with dashboards that allow students to control how suggestions appear and when they want additional help. Academics, meanwhile, receive succinct insights instead of raw data, enabling them to identify knowledge gaps, refine teaching materials and offer targeted feedback to specific cohorts.
To keep experimentation purposeful, the university is piloting a suite of classroom tools that adapt in real time to learner needs, while maintaining a strong emphasis on clarity and academic integrity. Among the most widely tested are:
- Adaptive reading lists that reorder articles based on prior knowledge and course progress.
- AI-informed lab simulations that vary complexity depending on a student’s confidence and error patterns.
- Multilingual support bots that clarify instructions and key concepts for international students.
| Tool | Primary Benefit | Typical Use Case |
|---|---|---|
| Personalised Quiz Engine | Targets weak areas | Pre-exam revision sessions |
| Study Path Recommender | Curates next steps | Planning weekly study goals |
| Feedback Summariser | Clarifies markers’ comments | Post-assignment reflection |
Building ethical and transparent AI frameworks for teaching and assessment
At the heart of Queen Mary’s approach is a commitment to designing AI systems that earn trust rather than demand it. This means codifying clear principles for how algorithms are trained, evaluated and deployed in learning environments. Staff and students are invited to scrutinise and shape these systems through open consultations, pilot projects and transparent documentation of data sources and decision rules.To ground this in practice, faculties are beginning to attach “AI audit trails” to digital assessment tools, allowing learners to see how automated feedback is generated, what parameters were applied and where human moderation has been used. This visible chain of accountability is reinforced by institutional policies that set non-negotiable red lines: AI must support, not replace, expert judgement; student data must never be repurposed without explicit consent; and any automated flagging of performance is always subject to human review.
These principles are reflected in everyday teaching and assessment design,where academic staff collaborate with technologists,ethicists and student representatives.Working groups emphasise:
- Explainability – making AI-generated feedback readable and challengeable by students.
- Fairness-by-design – stress-testing tools across disciplines, backgrounds and disability profiles.
- Co-agency – positioning students as active partners, not passive data points.
| Focus Area | Ethical Safeguard | Teaching Impact |
|---|---|---|
| Automated marking | Mandatory human moderation | Reduces bias, builds confidence |
| Learning analytics | Opt-in, minimal data profiles | Supports targeted, not intrusive, support |
| Feedback tools | Student-visible AI logic | Encourages critical digital literacy |
Empowering faculty and students with AI literacy and practical training
At Queen Mary, artificial intelligence is treated less as a distant technology and more as a shared language that every academic and student can speak. Dedicated workshops, micro-courses and embedded curriculum sessions focus on demystifying algorithms, data ethics and prompt engineering while openly addressing bias, transparency and accountability. Faculty are supported with hands-on clinics that show how AI can streamline research workflows,enrich feedback and create inclusive learning materials,while students learn to critically evaluate AI outputs rather than accept them at face value. This emphasis on critical digital citizenship ensures that tools like generative models, adaptive tutors and automated marking aids are understood, interrogated and responsibly integrated into everyday academic practice.
Training is designed to be highly practical and discipline-aware, with collaborative labs where engineers, lawyers, medics and humanities scholars experiment side by side. Short-form learning experiences include:
- Studio-style labs where staff and students co-design AI-enhanced lesson plans.
- Drop-in “AI office hours” offering rapid support on tools, prompts and policy.
- Sandbox environments to safely test models on real coursework scenarios.
- Ethics forums that examine academic integrity and responsible innovation.
| Programme | Audience | Key Outcome |
|---|---|---|
| AI Teaching Lab | Academic staff | Redesign of modules with AI activities |
| Prompt Studio | Undergraduates | Stronger critical and creative prompting skills |
| Data & Ethics Clinic | Postgraduates | Ethical frameworks for research with AI |
Strategic partnerships and infrastructure to scale responsible AI innovation in education
At the heart of Queen Mary University of London’s AI ecosystem is a web of alliances that blends academic rigor with real-world experimentation. The university is working side by side with edtech start-ups, school consortia and public agencies to co-design tools that are not only innovative, but deeply aligned with data ethics and inclusion. These collaborations prioritise transparent algorithms, student data sovereignty and fairness audits built into every prototype. To keep governance agile, cross-functional steering groups bring together computer scientists, teachers, legal scholars and student representatives, ensuring that classroom pilots are tested against robust ethical benchmarks, not just performance metrics.
Scaling this vision requires more than clever code, so Queen Mary is investing in an infrastructure layer that treats responsible AI as a shared utility rather than a luxury add‑on.This includes federated data platforms, sandbox environments for low-risk experimentation and cloud-native pipelines that can be replicated by partner institutions. Within this framework, collaborators are supported through:
- Secure data labs for privacy-preserving research and model training
- Open APIs that let schools plug AI tools into existing learning platforms
- Faculty advancement hubs focused on AI literacy and pedagogical integration
- Impact observatories tracking long-term effects on equity and attainment
| Partner Type | Key Focus | Shared Outcome |
|---|---|---|
| Schools & Colleges | Curriculum-aligned AI pilots | Evidence-based teaching tools |
| Edtech Start-ups | Co-developed learning platforms | Scalable, ethical products |
| Public Sector | Policy and standards | Trusted AI adoption at scale |
The Way Forward
As artificial intelligence continues to redraw the boundaries of what is absolutely possible in the classroom, Queen Mary University of London is positioning itself not as a passive observer, but as an active architect of this new terrain. Its work underscores a critical point: AI in education is not simply about automating tasks,but about reimagining how students learn,how teachers teach,and how universities serve an increasingly diverse and data‑literate society.
The university’s researchers and practitioners are testing the limits of what responsible, human‑centred AI can deliver, from personalising learning journeys to making higher education more accessible and inclusive. Yet the questions they are grappling with – around ethics, bias, transparency and academic integrity – will resonate far beyond one institution.
What happens at Queen Mary in the coming years will offer an critically important barometer for universities everywhere. If its experiments succeed, they may provide a blueprint for integrating AI into education without sacrificing rigour, equity or trust.If they fall short, they will still yield lessons on where the technology must adapt to the realities of human learning, rather than the other way around.
For now, the university stands at a familiar crossroads in the history of education: a moment when a powerful new tool is at hand, and the task is to ensure that it serves scholarship rather than subsumes it. How Queen Mary – and the wider sector – navigates that challenge will help determine not only the future of AI in education,but the future of education itself.
