This workflow is for overseas-qualified dentists preparing for ORE Part 1 who use, or are weighing up, an AI tutor such as ORE Mastery to explain practice questions. It targets the two written computer-based papers, where extended matching questions and single best answers reward discriminative reasoning. The principal limitation is blunt: an AI tutor that simply hands over the answer trains recognition, not reasoning. These 25 prompts make the tutor examine you first.
What ORE Mastery offers for ORE Part 1 right now
Lead with the honest finding. As of 19 July 2026 I could not independently confirm a single, clearly branded product called "ORE Mastery" with a published question count, defined access period, price and documented AI-tutor feature set for ORE Part 1. Several adjacent tools are easy to mistake for it: a "Dental Mastery: ORE & LDS" mobile app (which offers pre-written explanations rather than a conversational tutor), ExamineDental (which advertises AI-style analytics), and a range of Moodle-style ORE banks with human tutor support and tiered access, roughly in the £60–£120 band for one to twenty-four months (vendor-reported, 19 July 2026). None maps cleanly onto the "ORE Mastery" name with a confirmable, conversational AI tutor.
That matters: the archetype assumed here — an AI tutor you can talk to — may not exist under that exact name. So before you subscribe to anything marketed this way, verify the following on the product's own store or product page:
- The live question count, and how much of it is genuinely ORE Part 1 versus shared LDS or MFDS content.
- The access period and the price for the plan you would actually buy.
- Whether it provides a real conversational AI tutor, or only pre-written explanations plus analytics.
- Which components it covers — the two written papers only, or also Part 2 preparation.
The good news: the 25 prompts below are tool-agnostic. They work with any ORE Part 1 AI tutor, a general clinical AI, or the iatroX Socratic Tutor. The workflow is the asset; the specific vendor is interchangeable.
ORE Part 1: the official shape of the exam
ORE Part 1 is the written stage of the General Dental Council's Overseas Registration Examination. It is delivered as two computer-based papers, each three hours long, built from extended matching questions and single best answers. Paper A covers dental science and human disease; Paper B covers clinical dentistry together with law, ethics, and health and safety. You must pass both papers, and candidates are currently allowed four attempts. Passing Part 1 is the gateway to the Part 2 OSCE-style clinical examination, which this workflow does not address. The GDC does not publish a fixed per-paper question count in an easily citable form, so verify the current structure and counts on gdc-uk.org.
One editorial point matters for dental revision: the UK medicines reference here is the Summary of Product Characteristics (SmPC) via the electronic medicines compendium (eMC), read alongside NICE, SDCEP and GDC standards. So when a prompt below says "cite the source", that is what good looks like — not a chatbot's undated memory.
The prompt taxonomy: six jobs, one rule
Good testing prompts do six jobs: probe the underlying mechanism; force you to name discriminating features; drill option elimination; verify the governing guideline and its jurisdiction; run counterfactuals that show when a rule flips; and test cold retrieval. The single rule that ties them together is never to ask "what is the answer?" Instead, you commit to an answer and a rationale, then ask the tutor to test whether your reasoning holds. That inversion is the entire point.
Twenty-five prompts, grouped by the six error types they fix
1. Mechanism gaps — you knew the fact, not the "why".
- "I answered [option]. Before you say if I am right, ask me to explain the mechanism of [condition or drug action] in one paragraph, then critique what I missed."
- "Give me the single physiological or pathological step I have most likely skipped between [cause] and [clinical sign], and quiz me on it."
- "Turn this item into a 'why' chain: ask me three successive 'why' questions about [topic] and confirm the answer only after the third."
- "Explain how [drug] produces [effect] at receptor or tissue level, then ask me to predict what happens if [variable] changes."
2. Discriminating-feature confusion — the look-alikes.
- "List the two or three conditions or lesions most often confused with [diagnosis], then ask me for the one discriminating feature of each. Do not give me the features."
- "Draw a blank two-column contrast of [diagnosis A] versus [diagnosis B] and make me fill in the cells before you comment."
- "Ask me which single feature, if present, would move the diagnosis from [A] to [B], then mark my answer."
- "Quiz me on the radiographic appearance that separates [lesion A] from [lesion B] before revealing anything."
3. Option-elimination failures — extended-matching discipline.
- "For this option list, ask me to eliminate each wrong option and state why, one at a time, before I commit."
- "Give me one plausible distractor for [topic] and ask me to explain why it is wrong — not why the answer is right."
- "Ask me to rank all five options from most to least likely and justify the ranking, then challenge my weakest justification."
- "Present a stem where two options are defensible, then ask me which single feature makes one better."
4. Guideline and standard verification — anchored to the UK.
- "State the current UK or GDC standard for [topic — infection control, radiography justification, consent], cite the specific guidance and its date, then ask me to apply it to a new scenario."
- "Ask me what the GDC's Standards for the Dental Team require in [situation] before you comment."
- "For any medicine in this item, ask me to check the dose or indication against the SmPC/eMC, and tell me only whether I have it right."
- "Quiz me on the UK decontamination or health-and-safety principle relevant to [scenario], and correct only my errors."
5. Counterfactuals — testing where the rule breaks.
- "Change one variable in this stem — [age, medical history, drug, or timing] — and ask me whether the answer still holds and why."
- "Give me the version of this scenario where the 'obvious' answer becomes wrong, and ask me to name the tipping point."
- "Ask me to state the one contraindication or safeguarding flag that would override the standard management here."
- "Pose the same clinical problem in a medically compromised patient and quiz me on what changes."
6. Retrieval testing — cold recall, not recognition.
- "Close the explanation. Ask me to reproduce the key rule for [topic] from memory, then compare it with the source."
- "Give me three short free-recall prompts on [topic] with no options, and grade my precision."
- "Ask me tomorrow's version of this item: same concept, new stem, and mark whether I transferred the rule."
- "Test me on the numeric threshold or classification I am most likely to forget, then ask me to explain why the threshold exists."
- "Summarise nothing. Instead, ask me to teach [topic] back to you in five sentences and flag the sentence that is weakest."
The anti-answer-leak rule
Before you open any tutor, write down your chosen option and a one-line rationale. Only then paste the item into the tutor with one of the prompts above. The commitment step is what converts a passive explanation into a test. If you skip it, the tutor's fluent answer will feel like understanding while measuring nothing. This is the same failure mode we describe in the iatroX pillar on auditing an AI tutor: answer leakage inflates confidence without building recall.
A reusable jurisdiction check
Paste this after any answer: "State plainly whether this answer reflects UK dental practice and the ORE Part 1 jurisdiction, name the specific guideline or statute you are relying on, and give its date. If you are not certain it is current UK guidance, say so." A tutor trained largely on US or mixed content will often give plausible but jurisdictionally wrong answers on consent, safeguarding, radiography justification and prescribing — the exact areas Paper B tests.
Create a misconception record
For every genuine error, write four short lines: the incorrect rule you were using; the corrected rule; one transfer question that would catch the same slip in a new context; and a review date. Four lines, no more. A running misconception log turns scattered mistakes into a finite, revisable list — and it is the single most useful artefact to carry into the final fortnight.
Weekly verification sample
Once a week, take a small random set — say five — of the tutor's outputs and check them independently against the GDC, NICE, SDCEP or the SmPC/eMC. Log any discrepancy. If the tutor is wrong or out of date more than about one time in twenty on your sample, downgrade it from "source" to "sparring partner" and confirm every factual claim before you trust it.
Worked example: a seven-day plan
Assume an overseas-qualified dentist eight weeks out, using an ORE bank and an AI tutor for explanations, with iatroX for unseen transfer practice — a manual loop, with no proprietary-algorithm claims.
- Day 1: 30 mixed Paper A items in your bank. Log every miss into the misconception record.
- Day 2: Re-run only the missed concepts through the tutor using prompts 1–8. Commit first, then test.
- Day 3: 30 Paper B items (law, ethics, health and safety). Apply the jurisdiction check to every governance item.
- Day 4: Guideline-verification prompts (13–16) on the day-3 misses; confirm each against a primary UK source.
- Day 5: A fresh, unseen timed block in iatroX across both papers — no filtering, no assistance — to measure transfer rather than recognition.
- Day 6: Counterfactual and retrieval prompts (17–25) on the week's weakest three topics.
- Day 7: Weekly verification sample plus a 20-minute review of the misconception record. Set next week's floors.
The division of labour is deliberate: the bank supplies volume, the tutor supplies interrogation, and iatroX supplies the clean, unseen measurement that tells you whether any of it stuck.
Decision checklist: continue, supplement, switch or stop
- Continue if your unseen first-attempt score is rising and your misconception log is shrinking week on week.
- Supplement (add a second, unseen bank) if your bank score is high but transfer to fresh items lags — the classic recognition trap.
- Switch tutors if the weekly verification sample shows repeated jurisdiction or currency errors you cannot correct with prompting.
- Stop using the tutor for new learning once you are inside two weeks and reasoning is stable; shift entirely to timed, mixed, unseen blocks.
Make each call on a measurable gap, not on novelty or on how much you have already paid.
Bottom line
There may be no single confirmable product called "ORE Mastery", but the underlying question stands: can you make an AI tutor test your reasoning rather than feed you answers? These 25 prompts do exactly that, whichever tool you point them at. Keep the bank for volume, the tutor for interrogation, and reserve a clean, unseen set for measurement — because a percentage inside a bank you have already seen is not your exam score.
FAQ
Is ORE Mastery enough for ORE Part 1 on its own? Treat this cautiously: as of 19 July 2026 there is no clearly branded "ORE Mastery" product I could verify with a published question count and defined AI feature set. Any single ORE tool, whatever its name, is unlikely to suffice alone — you also need timed practice on unseen items, primary UK guidance for governance and prescribing, and enough volume to avoid memorising a finite bank. Confirm the product's scope on its own page before relying on it.
Which ORE Part 1 component does ORE Mastery not reproduce well? An AI tutor is weakest exactly where Paper B is strongest: UK-specific law, ethics, consent, safeguarding and health and safety, where answers hinge on current GDC standards and UK statute rather than clinical reasoning. Generalist models drift toward non-UK norms here, so any tutor's governance answers need the jurisdiction check and independent verification against primary sources.
How should I verify ORE Mastery AI answers for ORE Part 1? Run the weekly verification sample: take five random outputs, check each against the GDC, NICE, SDCEP or the SmPC/eMC, and log any error. Add the jurisdiction prompt to every governance or prescribing answer so the tutor must name and date its source. If it is wrong more than roughly one time in twenty, demote it to a sparring partner and verify every claim.
When should I stop using ORE Mastery and move to mixed mocks? Move once your unseen, first-attempt score across both papers is stable and your misconception log has stopped growing — typically the final two to three weeks. At that point, filtered single-topic practice yields little, and the binding constraint becomes pacing and stamina across a full three-hour paper. Switch to timed, mixed, unseen blocks with no assistance.
How should I combine ORE Mastery with iatroX without duplicating practice? Give each tool one job. Use the AI tutor to interrogate items you have already attempted, and reserve iatroX for fresh, unseen, timed blocks that measure transfer — never re-feed a competitor's item into iatroX or vice versa. That separation, described in the iatroX two-Q-bank rule, keeps your calibration honest: one bank to learn on, one clean bank to measure on.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; product figures are vendor-reported and change without notice, so verify counts, prices and features on the vendor's own page before purchase. Disclosure: iatroX operates a competing ORE Part 1 question bank and Socratic Tutor; this article confines iatroX to the unseen-measurement and transfer-practice role that a single AI tutor does not claim to fill, and it is deliberately honest that no single "ORE Mastery" product could be confirmed. Corrections are welcome via the feedback route on iatrox.com. References: General Dental Council ORE pages (gdc-uk.org); GDC Standards for the Dental Team; NICE, SDCEP and the electronic medicines compendium (medicines.org.uk) for the SmPC; the iatroX guide to auditing an AI medical-exam tutor and "Your Q-Bank Percentage Is Not Your Exam Score"; iatroX ORE Part 1 bank and comparison hub.
Open a missed ORE Part 1 item in the iatroX Socratic Tutor →