Bringing Generative AI into Higher Education: The ATLAS Approach
Bringing Generative AI into Higher Education: The ATLAS Approach
Generative AI has quickly become part of university life. Students are already using AI tools to search for information, organise ideas and support their writing, while teachers are experimenting with them to prepare learning materials, create activities and give feedback.
But using AI is one thing. Using it well in Higher Education is another.
Universities need to understand where these tools can genuinely improve teaching and student support, where they may create new challenges, and what teachers and staff need in order to use them responsibly. ATLAS responds to this need by combining practical experimentation, training and evaluation.
This is the challenge that ATLAS – Assisting Teaching and Learning with AI-based Solutions is addressing.
ATLAS is a 36-month Erasmus+ Partnership (2025–2028), bringing together 10 organisations, including eight universities, and 14 associated partners from 12 countries. Its aim is to help Higher Education institutions integrate Generative AI in a responsible, effective and pedagogically meaningful way.
Rather than focusing only on what AI could do, ATLAS asks a more practical question: what actually works in a university setting?
Could AI help teachers develop course materials or create assessments? Could it support more personalised learning? Could a chatbot answer routine student questions? Could AI provide timely feedback or support students during their thesis?
To explore these possibilities, ATLAS follows a use-case-driven approach. The project will develop and validate 60 Generative AI Use Cases for teaching and tutoring, covering areas ranging from content creation and assessment to student-information chatbots, academic support and feedback.
A central part of the project is experimentation in real educational contexts. Teaching and non-teaching staff, together with students, will test the Use Cases in existing courses and programmes across different disciplines and institutions. This makes it possible to understand not only whether a tool works, but also when it is useful, how it needs to be adapted and what challenges emerge in practice.
Starting in September, field experimentation activities with students and instructors will allow partners to test and further refine the solutions developed so far.
The process is iterative. Initial Use Cases are tested, reviewed, and then updated in a second cycle, allowing the project to build on evidence gathered in real settings and respond to the rapid evolution of Generative AI tools.
ATLAS is therefore not about introducing AI simply because it is new. The project also looks at its wider implications, considering pedagogical effectiveness alongside ethical and legal aspects. A common framework and testing protocols will help partners compare results across different universities and learning environments.
Generative AI is developing quickly, and there is unlikely to be one solution that works everywhere. What matters is testing different approaches, learning from experience, and understanding both the possibilities and the limits of these technologies.
Because bringing AI into universities is not simply about having access to new tools. It is about understanding when they are useful, how they should be used, and what they can really add to teaching and learning.
