PLAB is the gateway exam for international medical graduates entering UK practice, and its two parts test different transitions. PLAB 1 tests whether your medical knowledge has become UK medical knowledge: management per NICE, prescribing per UK convention, thresholds per UK screening. PLAB 2 tests whether your consultation style has become a UK consultation: structured, shared, safety-netted. AI tools now help substantially with both, provided you match tool to transition.
PLAB 1: making your knowledge British
The written paper punishes candidates who revise from internationally excellent but UK-inaccurate sources. That is the specific argument for question practice grounded in UK guidance: iatroX's PLAB 1 bank is curated against UK practice, its clinical answers cite NICE, CKS, SIGN and SmPC sources directly, and the adaptive engine plus spaced repetition schedule your weaknesses rather than your comfort. The Socratic Tutor matters more than usual for IMGs, because many PLAB 1 errors are not knowledge gaps but jurisdiction gaps, and a tutor that asks why you chose the other antibiotic surfaces exactly that. The bank sits in the £29 monthly or £99 annual tier, with samples free, and the same subscription later covers the MSRA and Royal College exams that follow UK registration.
A caution that applies platform-wide: US-centred AI explanations, however polished, are a systematic risk for this exam. Epinephrine dosing conventions, US screening ages and American first-line choices are precisely the distractors PLAB 1 writes. Whatever tool you use for explanation, verify management claims against UK guidance before they enter your long-term memory.
PLAB 2: making your consultation British
Geeky Medics is the standout here, and its own resources are explicitly used by PLAB 2 candidates: OSCE stations across history, examination, counselling and procedures, around 900 virtual patients to interview, and AI examiner feedback against mark schemes, with native apps for practice anywhere. The skills PLAB 2 scores, structured gathering, shared decision-making, safety-netting, land only through rehearsal with feedback, and rehearsing against an AI patient removes the practice-partner bottleneck that IMG candidates preparing abroad feel hardest.
ChatGPT can improvise extra patient role-play for free between structured sessions; it lacks mark schemes and calibration, so treat it as volume, not assessment.
A combined twelve-week shape
Weeks 1 to 6, PLAB 1 focus: daily adaptive question sets, Tutor on every miss, one timed block weekly, guideline reading only where errors cluster. Weeks 7 to 8: sit PLAB 1; begin light station work. Weeks 9 to 12, PLAB 2 focus: daily virtual-patient encounters on rotating station families, examiner feedback mined for knowledge gaps, those gaps closed with targeted bank questions and cited guideline answers, then the same stations re-run. Compress proportionally for an eight-week runway; the sequence holds.
The unifying principle: PLAB rewards candidates who train UK-specific knowledge and UK-style consultation as two deliberate projects. The AI stack that mirrors that split, curated UK-grounded questions plus virtual-patient rehearsal, is the one that shortens the distance.
Frequently asked questions
Is ChatGPT safe to rely on for PLAB preparation?
As an explainer of concepts, yes; as a source of UK management facts, no, and for this exam specifically the risk is systematic rather than random. General models default toward US conventions precisely where PLAB 1 writes its discriminators, so every management claim needs verifying against UK guidance before it enters your memory. The efficient version of that discipline is doing the verification once, through a tool whose answers already cite NICE and CKS directly.
How long do most candidates need?
Working IMGs commonly report three to four months of consistent part-time preparation for PLAB 1 and six to eight focused weeks for PLAB 2, with wide variance by baseline and hours available. The better planning unit is volume: a full pass of a curated PLAB 1 bank with misses converted, and a full rotation of station families with at least two debriefed cycles each. Adaptive scheduling compresses both by spending your hours on your actual weaknesses.
Which UK sources should anchor my preparation?
NICE guidance and CKS for management, SIGN where it leads, SmPC information for prescribing detail, and GMC material for the professionalism and ethics themes both parts test. The practical habit matters more than the list: whenever any AI tool explains a management decision, make it show which of these it stands on.
Does PLAB preparation transfer to what follows?
More than candidates expect. The UK-guideline reflex, the consultation structure and the retrieval habit all roll directly into foundation work and the MSRA, and on a multi-exam platform the account itself carries forward, which is a quiet argument against buying PLAB-only tools that expire on results day.
Is voice practice worth it for PLAB 2 specifically?
Disproportionately so, because the exam assesses consultation in spoken English under time pressure, and for many candidates the binding constraint is fluency rather than knowledge. Voice-mode virtual patients convert reading-speed preparation into speaking-speed performance, which is precisely the gap PLAB 2 examiners describe; schedule voice reps from the start of the PLAB 2 phase rather than saving them for the end.