For a new university student, the first few weeks can feel like being handed a very long list of things to work out.
There are forms to complete, accounts to activate, timetables to find and systems to access. Then come the questions that are harder to answer. Who do I speak to if I am struggling? Where do I find financial support? What happens if I need to change something on my course?
A surprising amount of this happens online, often before a student has attended their first lecture.
That makes the start of university an interesting test for artificial intelligence. If AI is going to make a meaningful difference in higher education, helping students find their way through these everyday interactions seems like a good place to start.
The opportunity is not simply to give students another chatbot. It is to make the university easier to understand and easier to access, while giving staff better information about the people they support.
Students don’t see the university the way staff do
Universities are complicated organisations. Admissions, registry, academic departments, finance, IT and student services all have different responsibilities, and many rely on separate systems to manage them.
That makes sense internally. Students experience something rather different.
A student with a question about their enrolment doesn’t necessarily know which department owns the answer. If they need help with their finances or wellbeing, they are unlikely to care which database contains the relevant information.
They just want to know what to do next.
This is one of the areas where AI could be particularly useful. A student should be able to ask a straightforward question in plain language and receive an answer that takes account of their circumstances, rather than being sent on a search through several university websites.
That sounds simple. In practice, it requires quite a lot of work behind the scenes.
AI is only as useful as the information behind it
There has been plenty of discussion about the potential of generative AI to answer student questions. Less attention is sometimes given to the less glamorous issue of where those answers come from.
A university might have the information a student needs, but it could be spread across its student record system, learning platform, website, policy documents and individual departmental systems.
Putting an AI assistant in front of all that information does not automatically make it useful.
If the information is inconsistent or out of date, students can get an answer that sounds convincing but isn’t right. In higher education, where a question might relate to an assessment deadline, funding, enrolment or a student’s support needs, that’s not a small problem.
Universities therefore need to think carefully about the information AI can access and the context in which it uses it.
The better the underlying information, the more useful AI can be.
Context is where things get interesting
A student doesn’t usually ask a question in isolation.
“Can I change my module?” might depend on what course they are studying, where they are in the academic year and what modules they have already taken.
“Why haven’t I received my funding?” could involve information held by several different parts of the university.
This is where AI has the potential to go beyond simply retrieving information.
With appropriate access to relevant data, an AI service could help bring together the information needed to understand a student’s question and guide them towards the right next step.
There are obvious boundaries here. Universities need to be careful about privacy, permissions, accuracy and how automated systems make use of sensitive information. AI should not be making important decisions about students simply because the technology makes it possible.
But there is a useful middle ground between a basic FAQ and full automation.
AI can help students navigate complicated services while leaving decisions and more sensitive conversations with the people responsible for them.
The best use of AI may be getting students to a person
One of the assumptions around AI in education is that its value comes from reducing the number of interactions students have with staff.
I think that is too narrow.
Sometimes a student really does need a quick answer. If AI can provide it at 10pm without requiring someone to search through six university webpages, that’s useful.
But there are plenty of situations where the right outcome is a conversation with a person.
A student may be dealing with financial difficulty or finding the transition to university harder than expected. An AI system might help identify the relevant support service or give staff useful context, but it should not be mistaken for the support itself.
This is particularly important as universities explore more personalised uses of AI. The technology can help remove some of the administrative work around student support, giving staff more time to focus on the situations where judgement and empathy matter.
That is a more useful measure of progress than simply counting how many queries an AI system has answered.
AI could help universities spot problems earlier
There is another opportunity that sits slightly further behind the student interface.
Universities already collect large amounts of information about their students. Attendance, engagement, assessment activity, academic records and interactions with support services can all provide pieces of a much bigger picture.
AI could help staff make sense of those signals.
Imagine a student who has missed several appointments, hasn’t completed an important administrative task and has stopped engaging with course information. None of those things necessarily means there is a problem. Taken together, they might be enough to prompt someone to check in.
AI should not decide that a student is struggling or determine what support they receive. It can help staff notice patterns that might otherwise be difficult to spot across disconnected systems.
The conversation that follows still needs a person.
For some students, that earlier conversation could make a meaningful difference. Catching a problem while it is still manageable gives a university more options than waiting until the student has reached crisis point.
There is a practical benefit for staff too
AI in higher education is often discussed in terms of teaching and learning. The operational side deserves just as much attention.
University staff deal with thousands of routine questions, particularly around the start of the academic year. Many are predictable and involve information that already exists somewhere within the institution.
If AI can handle some of those questions accurately, staff can spend less time repeating the same information.
Across a large university, it can add up to a significant amount of time.
More importantly, the time saved can be used elsewhere.
A member of student services who isn’t spending part of their day answering basic process questions has more capacity for a student with a complicated problem. An administrator who isn’t manually moving information between systems can spend more time resolving an issue that genuinely needs their attention.
The value of AI is therefore partly about what it allows people to stop doing.
Universities need to resist the urge to add AI everywhere
There is a danger in the current enthusiasm around AI that every problem starts to look like an AI problem. It isn’t.
Some university processes simply need to be redesigned. Some information needs to be brought together. Some websites need to be easier to use. And some tasks are better handled by a person.
AI works best when there is a clear reason for using it.
For universities, that means looking at the student journey and identifying where people regularly get stuck. Where are students searching for information? Which questions keep coming into support teams? Where are staff having to piece together information manually?
Those are the places where AI might have something useful to contribute.
The technology should earn its place.
Making AI useful starts with the basics
Higher education has a genuine opportunity to use AI to improve the way students interact with their university.
But the most valuable applications may not be the most visible ones.
They could be the systems that quietly help a student find the right answer, give a member of staff better context before a conversation or flag that someone may need support earlier.
That requires universities to get some basic things right. Information needs to be accurate and accessible. Systems need to work together where there is a genuine benefit in doing so. Staff need to understand where AI is being used and where human judgement remains essential.
None of that is particularly flashy.
It is, however, what will determine whether AI becomes a useful part of university life or simply another technology students and staff have to navigate.
The best digital experiences tend to disappear into the background. AI has the potential to do the same in higher education – helping students get where they need to go, helping staff understand when they are needed and taking care of some of the routine work along the way.
For a student starting university, that could make the first few weeks feel a little less complicated.
And that would be a worthwhile place for AI to make its mark.
