The university in the age of intelligence

Intelligence is becoming scalable in a way that would have sounded like science fiction a few years ago. You can see it in small, almost funny moments, such as how quickly workplace emails became more polished and expressive when AI tools became widely available. Presentations that once took days can be put together in short order and a first cut of a consultant-grade business plan can appear in a jiffy. AI is helping people make sense of an avalanche of data, spot patterns and draw insights from volumes of information that were previously difficult to wrangle.  And increasingly, AI-enabled automation is beginning to take on routine work that previously required people to move information, interpret it and decide what should happen next.

Clever emails, faster presentations, data analytics and automation on steroids and business plans before lunch are useful, but for universities, the bigger opportunity is what broader AI enablement makes possible across the institution.  The changing nature of academic research provides a window into the possibilities, with AI increasingly becoming part of the research process.  It is helping researchers navigate papers and analyse datasets at a scale difficult for an individual or small team to manage unaided, while identifying possible connections and exploring research pathways faster. The objective is to reduce the grunt work around discovery, giving researchers more time to question findings, exercise judgement and develop original ideas.

The opportunity extends much further though. Universities face a demanding set of expectations and difficult trade-offs. They are under pressure to extend access, improve teaching and research, provide better student support and keep costs under control. Traditionally, doing more has required more people, more infrastructure and more money.

AI should alter that equation. Certain forms of expertise and support can now reach far more people without costs rising at the same rate.  Making this work still requires investment in systems, data and people, but it gives universities a chance to increase what they can achieve with the resources available.  

Start with the student experience. Early university chatbots mostly helped students find their way around, telling them how to register, which forms to complete and whom to contact. Some universities have now extended these bots into more active student support, using reminders and gentle nudges to help students juggle assignments, deadlines and other milestones across multiple courses.  A further step would be to connect signals that currently sit apart, such as whether a student stops attending lectures, disengages from a learning platform or no longer arrives for meals. Each observation might mean little on its own. Together, they could prompt someone to check in. Students should obviously understand how their information is used. Used responsibly, these signals could help a university notice a struggling student sooner.

In teaching, AI tutors offer the prospect of individual attention at a scale few universities could otherwise afford. AI-enabled tutoring systems can increasingly provide personalised support, responding to where a student is struggling in a style and at the pace at which they are progressing. Sal Khan has described the ambition in compelling terms, as the potential to give every student on earth a personal AI tutor and every teacher an AI teaching assistant. The bigger idea is individual attention at a level that conventional economics has never really allowed.

Then there is the physical university itself. Imagine energy and water systems, HVAC, buildings and other infrastructure increasingly equipped with sensors, with that information brought together and monitored holistically rather than each system being viewed in isolation. This creates a much richer understanding of how the campus behaves.  This could reveal where energy is being wasted, how space can be better utilised, where demand is changing and where equipment may require attention before it breaks down. This shifts campus management from simply reacting to what has happened towards anticipating what is likely to happen, while reducing costs and the risk of disruption.

Universities have an unusual advantage in tackling these challenges. They already contain much of the expertise required in the form of a concentration of some of the brightest minds across multiple disciplines, with skills and problem-solving capability that can be directed to solve some of the challenges of the institution itself. The campus can become a place to experiment, solve problems, test ideas, learn and improve. Students and academics do not simply need to be users of the Intelligent University. They can help create it.

Part of getting there will undoubtedly involve connecting systems and things that today often sit apart. Universities generate enormous amounts of information across teaching, research, student services, finance, administration, facilities and the physical campus. Bringing more of that data and intelligence together creates a more complete view of the institution and how its different parts interact. It makes it easier to see patterns, trade-offs and opportunities across teaching, research, operations and the student experience.  

That leads to a more ambitious possibility. A leadership dashboard could develop from a record of past performance into a way of exploring future choices.  What happens if enrolments grow in one faculty? How much additional capacity can the existing campus absorb? What does that mean for lecture space, residences, energy, student support and cost?  Which interventions might improve student outcomes? What happens if resources are shifted from one area to another?  Instead of seeing the consequences of important decisions after the fact, increasingly sophisticated models create the possibility of exploring those consequences beforehand. 

The Intelligent University would take shape through those choices, with an integrated perspective on where to invest, what to connect, when to intervene and where human judgement matters most. Its success would be measured in better learning, stronger research, broader access and more effective use of resources.

After all, the purpose is not just to create more intelligence. It is to create more human potential.