Loading…
Say your answer out loud first, then jot down the key points you made. Writing it helps you spot gaps — and it's what the AI feedback reads.
0 words
Balanced, informed thinking about technology in medicine, including benefits, risks, ethics and what remains distinctively human in the doctor's role.
A structure to build your own answer on — not a script.
Where AI helps now
Image analysis (e.g. radiology, retinal scans), triage, documentation and administration, predicting deterioration.
Benefits
Speed, consistency, freeing clinician time, access.
Risks
Bias from training data, errors and over-reliance, accountability, privacy, explainability, digital exclusion.
What remains human
Judgement under uncertainty, integrating context and values, communication, trust, accountability.
Conclude
Changes rather than replaces the role; doctors need to use it critically.
A strong but not perfect answer. Read the notes to see why it works — then make your own version with your experiences.
I don't think AI will replace doctors, but I think it will change what doctors spend their time doing1.
It's already useful in areas that involve recognising patterns in large amounts of data — reading scans, screening retinal images, flagging patients whose observations suggest they're deteriorating. Tools that write up clinical notes from a consultation could give doctors back time with patients, and the NHS's 10 Year Health Plan puts a lot of weight on this kind of digital shift.
But there are real risks. AI systems learn from existing data, so if that data under-represents certain groups — for example, skin conditions in darker skin2 — the system may perform worse for them and widen inequalities. There's the risk of over-reliance, where clinicians stop questioning outputs. And there are hard questions about accountability when something goes wrong, and about privacy.
What I think stays human is the part of medicine that sits around the data: understanding what matters to a particular patient, weighing uncertain options with them, communicating bad news, taking responsibility for a decision. A system might tell you the probability of a diagnosis; it doesn't sit with the family deciding what to do3.
So I'd see AI as a powerful tool that doctors will need to understand well enough to use critically — knowing its limits, checking it against the patient in front of them, and speaking up when it's wrong.
Real interviews are won and lost here. Answer each one aloud.