AI literacy is no longer enough: Why universities need institutional AI capability

Why universities need to move beyond AI literacy

One of the key themes discussed at the Summit in Budapest was the evolution from AI literacy to awareness, fluency and, ultimately, intentionality. While AI literacy is about understanding the fundamentals of how AI works, AI awareness involves recognising its risks, ethical implications and broader impact. AI fluency is the ability to confidently and effectively use AI tools, prompts, assistants and agents in everyday work. Together, these capabilities lay the foundation for more intentional and strategic use of AI.

AI intentionality pushes users to ask a more fundamental question. As Dr Michelle Sisto, Professor of AI and Decision Sciences at EDHEC Business School, put it: “Should I use AI, depending on what my objectives and goals are?”

The ability to use AI is not the same as knowing when to use it. Nor does it guarantee an understanding of when its outputs can be trusted, where its limitations lie, or when human judgement should take precedence.

Crucially, these capabilities cannot be developed through students alone. Sisto argued that higher education must address these questions “not just [for] students and participants, but across the entire spectrum of stakeholders”.

This broadens the challenge beyond student AI literacy. Building AI readiness requires a coordinated, institution-wide approach, with educators, leaders and professional staff aligned on how, when and why AI should be used.

What does an AI-ready university look like?

An AI-ready university is not necessarily the institution using the most AI. Instead, it is one capable of making informed decisions about where AI creates value, where it introduces risk and where human expertise remains essential.

Cerys Hutton, Head of Warwick Digital Academy at the University of Warwick, highlighted the scale of the organisational change required, arguing that there are “skills development changes, not just from the student perspective, but from the staff perspective”. This means universities need to consider the AI capabilities of academics, professional services staff and institutional leaders alongside those of students.

Being AI-ready also requires universities to become more agile. Hutton identified “agility, adaptability, and kind of being able to respond to the urgent challenge” as increasingly important, alongside “having the structures in place for legal and rapid operation”.

The goal should not be AI adoption for its own sake, but rather, universities need governance, processes and skills that enable them to make deliberate, evidence-based decisions about AI while keeping pace with evolving technology.

AI capability is a governance and culture challenge

Effective AI governance requires more than publishing policies. Institutions need the structures, skills and culture to put those policies into practice.

During conversations on responsible AI at the Summit, audience members were asked whether their institution had an AI strategy extending at least five years into the future. Only two hands were raised – highlighting the gap between growing AI adoption and long-term institutional planning.

Even where policies exist, awareness cannot be assumed. Hutton shared that a recent survey at Warwick found that "50% of people weren't aware [they] had the information security policy". This highlights an important distinction between having AI governance and having AI capability. Policies need to be understood, communicated and embedded into everyday decision-making.

Culture plays an equally important role. Hutton described a "polarisation of attitudes" between those reluctant to engage with AI and those eager to embrace it. The risk is that institutions become "stuck in the middle, where nobody takes the decision."

For universities, understanding their current level of AI readiness is therefore an important starting point. The QS AI Capability Framework helps institutions understand what effective AI adoption looks like across governance, teaching, research and operations, while the QS AI Capability Self-Assessment and Analyst-Led Assessment provide ways to measure current maturity, identify capability gaps and prioritise action.

Measurement, however, is only part of the challenge. Building capability also requires opportunities to learn, experiment and collaborate. Through the Responsible AI Consortium (RAIC), institutions can engage in peer learning, collaborative research and experiential projects, while exploring and piloting AI technologies alongside other universities.

Together, these approaches can help institutions move beyond isolated AI initiatives towards a more coordinated model of adoption – one grounded in effective governance, shared responsibility and responsible experimentation.

Responsible AI depends on trust

Institutional AI readiness cannot be treated solely as a technology challenge because AI is also reshaping relationships within universities.  

Dr Csaba Csaki, Dean for Artificial Intelligence at Corvinus University of Budapest, argued that amid rapid technological change, "we kept forgetting about humans". He highlighted a growing trust challenge, with students questioning the use of AI in assessment and lecturers concerned about AI-assisted assignments. His conclusion: "We have to re-establish the trust within us."

Trust depends on transparency. Students and staff need to understand not only what AI can do, but where it is being used, why it is being used and what safeguards exist around those decisions.

Discussions in the creative arts offered a useful model. Rather than attempting to prohibit AI, educators described approaches centred on experimentation, transparency, ethical guidance and critical reflection. The emphasis was not on maximising AI use, but on helping students and staff develop the judgement needed to determine when AI adds value and when human expertise should take the lead.

From AI literacy to institutional AI readiness

The first phase of higher education's response to generative AI focused on access. The next focused on AI literacy. The challenge now is institutional AI readiness.

As AI continues to reshape both higher education and the labour market, universities face a dual challenge: preparing students for an AI-enabled world while ensuring their own institutions are equipped to navigate the same transformation.

‍