How do user experience and user interface differ?
The Harvard Chan AI Glossary
In our classes, we often pause to explain the ideas behind AI and the language used to describe it. These posts grew out of those conversations. They’re for anyone who would like to understand AI better, with no technical background assumed.
The user interface, or UI, is what a person sees and interacts with: buttons, forms, menus, messages and other controls. User experience, or UX, is broader. It concerns whether the overall interaction is understandable, supports the person’s task and fits the circumstances in which it is used.
Imagine an application that produces an AI summary. The UI includes the upload control and the panel showing the summary. The experience also includes understanding which document was used, waiting for the result, checking its claims and deciding what to do if it is wrong. A visually appealing panel can coexist with a frustrating or misleading experience.
A workflow is the sequence of work and handoffs through which a task gets completed. Much of it may happen outside the application. A person may have to search for a file, ask a colleague for access or copy the result into another system. Good UX considers those connections rather than treating the screen as the entire job.
Accessibility concerns whether people with different abilities and circumstances can use the product. Readable text, keyboard operation and understandable instructions are not merely aesthetic choices. They affect who can complete the task. A design that works for its creators is not necessarily usable by the intended audience.
AI adds specific questions. Does the person know whether a response is a draft or a completed action? Can they inspect the source? Can they correct the result without starting again? Does the system acknowledge missing information? The experience depends on these behaviors, not just the quality of the generated sentences.
Testing UX means observing representative people attempting realistic tasks. It is different from asking whether they like a screenshot. Watch for hesitation, workarounds, repeated attempts and misunderstandings. These observations can reveal a mismatch between the team’s assumptions and actual use.
For a clinic follow-up tool, the UI might be a clear list of patients and next actions. The experience also includes whether staff can trust its freshness, find the relevant record, reach the patient and record what happened without entering the same information twice. An attractive list could still increase work if those steps are disconnected. The dental-care discussion in class is a useful starting point: the visible dashboard matters because of the care pathway around it, not as a display of AI capability.
If you would like to find out more about AI and improve your leadership skills, consider the AI in Health Care Certificate of Specialization from the Advanced Learning Academy at Harvard T.H. Chan School of Public Health. You can take an individual course or work toward the full certificate.
AI is a fast-moving field. This explanation reflects our class discussions and related material at the time of writing. Terminology, capabilities and tools may change.
For suggested additions or corrections, please email me at tpanch@hsph.harvard.edu.
Dr. Trishan Panch
Program Director, AI in Health Care
Harvard T.H. Chan School of Public Health — Advanced Learning Academy