AI Levels Episode 5: AI on the front line of the student experience
It’s 11pm, and a student has a question about changing a module, meeting an application deadline or accessing support. They want to know what to do next. An AI assistant that provides clear, reliable guidance can help them move forward, even when university offices are closed.
That is one of the opportunities explored in Episode 5 of ICS.AI’s AI Levels podcast, “The Student Experience, AI on the front line of learning”.
Andy Logan looks at how AI can support enrolment, onboarding, advising and personalised learning, alongside the decisions institutions need to make about accuracy, fairness and human oversight. How can institutions make support more accessible while ensuring each student’s circumstances are understood?
Reliable support starts with the right information
For prospective and current students, navigating an institution can involve finding the right department, understanding an unfamiliar process or working out which form to complete.
An AI assistant can make those tasks easier. It can explain entry requirements, guide a student through a module change or help them find the relevant support service. By handling routine enquiries, it can also give admissions, registry and support teams more time for situations that need individual attention.
As Andy explains, those benefits depend on grounding: connecting the assistant to accurate, current, institution-specific information. A helpful tone offers little reassurance if the answer contains the wrong deadline or an outdated appeals process.
This is directly relevant to ICS.AI’s SMART: Front Door, which brings together AI-powered support across channels including phone, webchat and email, with round-the-clock multilingual assistance and escalation to staff when human input is needed. In education, the aim is to help students reach the information or service they need, with a clear route to personal support when their circumstances require it.
Understanding the decisions that affect students
The episode opens with the 2020 exam grading controversy. Following concerns about inconsistent and unfair outcomes, the Government and Ofqual announced a return to centre assessment grades, retaining calculated grades where they were higher.
Andy uses this example to explore a question that also matters for AI in education: can an institution explain an outcome to the student affected by it?
The episode’s AI Level Up segment introduces the “black box” problem, where the reasons behind a system’s output are difficult to understand.
That becomes particularly significant when a system influences a course recommendation, flags a student as at risk or shapes the opportunities presented to them. Institutions need to understand the evidence behind those outputs, recognise its limitations and ensure someone can review the outcome with the student.
Making personalisation meaningful
Personalised learning offers considerable promise. A student struggling with a particular concept could receive additional material, while someone who has demonstrated understanding could move ahead.
Delivering that experience requires careful educational design. Different learning pathways still need to prepare students for shared assessment criteria and qualifications. Early recommendations also deserve scrutiny, because a system may initially know very little about a new learner.
Andy explores the risk of treating patterns among similar students as a complete picture of an individual.
The same consideration applies to pastoral support. A commuting student, a carer or someone who prefers studying at home may spend less time on campus for reasons unrelated to their commitment to learning. Engagement data needs to be interpreted with that context in mind.
Used thoughtfully, a pattern can prompt a valuable conversation. Understanding what support would help requires attention to the person behind the data.
Giving students confidence in AI support
Transparency is a practical part of building a useful student experience.
Students should understand when they are interacting with AI, what information it can access and who they can contact if an answer is wrong. Where AI influences a consequential decision, an accountable member of staff should be able to review it and discuss it with them.
This gives students a way to challenge an assumption, explain their circumstances and seek appropriate help.
For institutional leaders, the episode offers a useful starting point: look at your AI services through a student’s eyes. Can they rely on the guidance? Can they understand why something has been recommended? Do they know where to turn when they need a person?
Listen to AI Levels Episode 5
Join Andy Logan for a closer look at student-facing AI, the tension between personalisation and standardisation, and the practical choices that help institutions earn students’ trust.
Comments