This audit is for AMC MCQ candidates weighing PassAMC and PassAMCQ's AI tutor as a study companion. It covers the written single-best-answer paper only, not the AMC Clinical Examination. The principal limitation is structural: any generative tutor is fluent and confident by design, so its explanations must be treated as a hypothesis to verify against Australian sources, not as settled fact. Used with that discipline, it accelerates learning; used credulously, it teaches errors persuasively.
What PassAMC and PassAMCQ offer for the AMC MCQ right now
One honesty point first. "PassAMC" and "PassAMCQ" appear across several linked web properties, and their advertised figures differ by page. Treat the brand as a family of products and confirm what you are actually buying on the specific checkout you use. The following are vendor-reported and last checked on 19 July 2026.
| Attribute | PassAMC / PassAMCQ (vendor-reported, 19 July 2026) |
|---|---|
| AI feature | A "PassAMC AI Virtual Tutor" described as "trained for AMC MCQ & Clinical success"; specific capabilities are not detailed |
| Question volume | Figures vary by property: passamcq.com states "1,600+"/"1,800+" MCQs; the passgp.au property states 3,500 MCQs plus 1,050 clinical cases; passamc.org states "over 2000" — verify on the page you buy from |
| Mock exams | passamcq.com lists Mock Exam 1 and 2 (100 items each); the passgp.au property references up to 7 MCQ mocks — verify current provision |
| Analytics | "Advanced progress tracking"; the passgp.au property references cohort comparison and "predictive analytics" |
| Localisation | passamcq.com references Therapeutic Guidelines (eTG), NPS MedicineWise and Royal Colleges guidance; authorship attributed to Australian doctors and former AMC/PESCI/PLAB examiners |
| Price | passamcq.com lists AUD 150 (1 month) to AUD 450 (12 months); passamc.org lists roughly USD 30–80; confirm current price and currency on the product page |
| Components supported | The AMC MCQ written paper (the AI tutor is also promoted for clinical prep) — it is not a substitute for the AMC Clinical Examination |
The AI Virtual Tutor is a real vendor-advertised feature, but the page does not detail how it is grounded, what model powers it, or whether its explanations are retrieval-linked to sources. That gap is precisely what this audit tests.
The exam you are actually preparing for
The AMC MCQ is 150 single-best-answer questions, one correct option of five, in one 3.5-hour computer-administered session at Pearson VUE. The AMC MCQ Specifications (V8, 9 September 2025) define the content: the disease process, clinical examination and diagnosis, investigation, and therapy and management, across Adult Health (Medicine ~30%, Surgery ~20%), Women's Health (~12.5%), Child Health (~12.5%), Mental Health (~12.5%) and Population Health and Ethics (~12.5%). Confirm the current percentages in the latest specification. The exam is the gateway to the AMC Clinical Examination. It is named a Computer Adaptive Test, but the AMC does not confirm an adaptive-difficulty algorithm — prepare for a fixed, whole-blueprint paper. Any AI answer that contradicts the specification loses to the specification, every time.
Testing methodology: a fixed rubric across six item types
Do not judge an AI tutor on one lucky answer. Put it through the same six representative item types every time, and score each against a published rubric so your assessment is repeatable rather than impressionistic:
- Straight recall — a factual item with one unambiguous answer.
- Diagnosis — a vignette requiring you to weigh a differential.
- Next investigation — where the trap is over- or under-investigating.
- Management — where the correct step is Australian-specific (drug choice, referral pathway, notification).
- Ethics or medico-legal — consent, capacity, mandatory reporting in the Australian context.
- A deliberately ambiguous item — to see whether the tutor acknowledges uncertainty or manufactures false confidence.
Score each answer on five axes: correctness, provenance (does it cite a source you can check), reasoning quality (does it explain why, not just what), jurisdiction (Australian, not UK or US), and calibration (does it flag its own uncertainty). A tutor that scores well on correctness but poorly on provenance and jurisdiction is a fluency engine, not a teacher.
Grounding audit: where do the answers come from?
The single most important question about any AI tutor is provenance. Read three or four explanations and ask: does the answer cite a checkable source — a named guideline, the eTG, the Australian Medicines Handbook — or is it uncited prose that merely sounds authoritative? Does it distinguish the vendor's own written explanation from model-generated commentary? You can inspect this without reproducing any copyrighted content: note whether a source is named and specific ("per the National Immunisation Program schedule") versus vague and appealing ("guidelines recommend"). Vague attribution is the tell that an answer is ungrounded. When PassAMC's tutor names a source, verify the claim in that source; when it does not, treat the claim as unverified until you have anchored it yourself.
Reasoning behaviour: teacher or answer-key?
A good tutor changes how you think; a poor one just hands you the letter. Assess four behaviours. Does it ask a useful diagnostic question before answering, or reveal the answer immediately and remove the productive struggle? Does it handle uncertainty — will it say "this is genuinely contested" — or does everything arrive with the same false certainty? Does it correct a false premise: if you assert something wrong in your prompt, does the tutor push back or agree agreeably? And does it explain the distractors — why the other four options are wrong — which is where most of the learning in single-best-answer questions actually lives? Answer leakage, where the tutor telegraphs the answer before you have committed, is the most common way an AI companion quietly destroys the value of a practice item.
Exam fidelity: does the advice fit the AMC?
Fluent medical prose is not the same as AMC-appropriate advice. Probe fidelity directly. Does the tutor respect Australian jurisdiction — Pharmaceutical Benefits Scheme realities, Australian referral pathways, notifiable-disease rules — rather than importing UK or US management? Does it use Australian terminology and reference Australian sources? Does its advice acknowledge the exam's time pressure, roughly 84 seconds per item, rather than modelling a leisurely clinical work-up? A tutor that recommends a UK or US first-line drug, or a formulary that does not apply in Australia, has failed the fidelity test regardless of how polished the explanation reads.
Failure modes to watch for
Every generative tutor exhibits a predictable set of failures. Learn to spot them:
- Hallucinated citations — a plausible-sounding reference to a guideline that does not say what is claimed, or does not exist.
- Overconfident wording — uniform certainty applied to genuinely uncertain questions.
- Outdated guidance — advice that lags the current eTG, immunisation schedule or screening programme.
- Answer leakage — revealing the answer before you have reasoned to it.
- Plausible-but-unexamined elaboration — extra detail that sounds educational but was never in the question and may be wrong.
None of these mean the tool is useless; they mean it is a draft to be checked.
Safe-use protocol: answer first, interrogate second, verify third
Impose a three-step routine on every item, and the tutor becomes an asset rather than a crutch:
- Answer first. Commit to your own answer before opening the tutor. This preserves the retrieval effort that makes practice work and prevents answer leakage.
- Interrogate second. Then ask the tutor to justify — high-value prompts include "why is option C wrong, specifically?", "what Australian guideline supports this management?", and "what would change this answer?" Make it defend its reasoning, not just state a letter.
- Verify third. For any management, drug or schedule detail, confirm it against the eTG, the Australian Medicines Handbook or the relevant Australian programme before you commit it to memory. If the tutor cannot name a checkable source, do not bank the fact.
A seven-day pattern for international graduates
Give PassAMC and PassAMCQ one job — explanation and content review through its tutor and bank — and give iatroX a separate job: unseen, mixed, timed transfer practice. No claim is made about either platform's internal algorithms.
| Day | PassAMC / PassAMCQ (learning job) | iatroX (measurement job) |
|---|---|---|
| Mon | 30 items; answer-first, then interrogate the tutor on distractors | — |
| Tue | Verify Monday's management points against eTG/AMH; 20 transfer items | — |
| Wed | 30 items in a weak discipline; log provenance failures | Unseen mixed 20-item timed block; first-attempt only |
| Thu | Ethics and Australian medico-legal set through the tutor | — |
| Fri | 30 mixed items; re-test corrected false premises | — |
| Sat | Review flagged explanations you had to correct | Unseen mixed 50-item timed block, scored once |
| Sun | Rest; plan next week from Saturday's gaps | — |
Decision checklist: continue, supplement, switch or stop
- Continue if the tutor's explanations are grounded, jurisdiction-correct and improving your first-attempt accuracy on unseen items.
- Supplement if you frequently have to correct its guidance, or if its scores outrun your independent mixed-block performance — add an unseen measurement source.
- Switch the primary tool if provenance is consistently vague, guidance repeatedly imports non-Australian management, or citations do not check out.
- Stop adding new questions and move to full mixed mocks once every discipline clears its floor on unseen items and pacing is under budget.
The bottom line
PassAMC and PassAMCQ's AI Virtual Tutor is a genuine, Australian-facing study aid, and its examiner-attributed authorship and eTG references are points in its favour. But it is still a generative tutor, which means its confidence is manufactured and its explanations are a hypothesis. The candidates who benefit are the ones who answer first, interrogate the reasoning, and verify every management detail against Australian sources. Fluency is not fidelity; make the tutor prove it.
Frequently asked questions
Is PassAMC and PassAMCQ enough for AMC MCQ on its own? It can be a strong core, given its Australian authorship and eTG-referenced explanations, but "enough" depends on independent verification of readiness. A tool whose questions and tutor you have already worked through cannot also serve as your unbiased measurement, so most candidates pair it with an unseen, mixed, timed source. Confirm the current question count on the specific PassAMC or PassAMCQ property you buy from, as the figures differ across their sites (vendor-reported, 19 July 2026).
Which AMC MCQ component does PassAMC and PassAMCQ not reproduce well? It targets the written single-best-answer paper and promotes the tutor for clinical prep, but it does not reproduce the AMC Clinical Examination itself — the observed-consultation stage cannot be replaced by an MCQ tutor. Its AI explanations also cannot reproduce the exam's most important discipline: independent, whole-blueprint reasoning under time pressure with no assistance available.
How should I verify PassAMC and PassAMCQ AI answers for AMC MCQ? Apply the answer-first, interrogate-second, verify-third routine. Commit your own answer before reading the explanation, ask the tutor to justify why each distractor is wrong and which Australian source supports its management, and then confirm any drug, schedule or pathway claim against Therapeutic Guidelines (eTG), the Australian Medicines Handbook or the relevant Australian programme. If it cannot name a checkable source, do not memorise the claim.
When should I stop using PassAMC and PassAMCQ and move to mixed mocks? Stop drilling explained single items once every discipline clears a first-attempt floor on unseen questions and your pacing sits near 84 seconds per item. Beyond that point, more explained practice adds diminishing returns; your remaining gains come from full-length, mixed, timed mocks that rehearse endurance and unaided decision-making across the whole blueprint.
How should I combine PassAMC and PassAMCQ with iatroX without duplicating practice? Assign distinct jobs and never share items. Use PassAMC and PassAMCQ for explanation and content review, and use iatroX solely for unseen, mixed, timed measurement. Because the iatroX items are independent of what its tutor has already walked you through, an iatroX block is a fair test of transfer — evidence that the tutor's teaching has actually stuck rather than a rehearsal of answers you have seen.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 19 July 2026; all PassAMC and PassAMCQ figures are vendor-reported and differ across the brand's web properties — verify the current AI-tutor capabilities, question count, price and currency on the specific product page before relying on them.
Disclosure: iatroX operates its own AMC MCQ question bank and Socratic Tutor and is therefore a competitor to PassAMC and PassAMCQ. This audit confines iatroX's role to the unseen, mixed, timed measurement job that a study tutor is not designed to serve, and makes no claim about any platform's internal algorithms. Corrections are welcome via the feedback route on iatrox.com.
References: AMC Computer Adaptive Test (CAT) MCQ Examination and AMC MCQ Specifications (V8, 9 September 2025), amc.org.au; PassAMC and PassAMCQ product pages (passamcq.com, passgp.au, passamc.org); Therapeutic Guidelines (eTG), the Australian Medicines Handbook and the National Immunisation Program for Australian management standards. Internal reading: how to audit an AI medical exam tutor, Your Q-Bank Percentage Is Not Your Exam Score and the iatroX comparison hub.
Answer a fresh AMC MCQ in iatroX and interrogate it before you read the explanation →