Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2221
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-23T18:35:10.215928Z
status: published
lens: p_space
sector: education
headline: The Differential That Never Gets Generated
dek: Residents increasingly ask large language models to name a lesion before diagnosing it themselves. The practice contributes a specific harm: never-skilling, a phenomenon in which reasoning is never acquired, and is distinct from 'deskilling' or 'model error'. Forming a differential first is necessary but insufficient. Departments should require documented differentials, supervise disagreement, and examine unassisted reasoning.
gain_title: (none)
problem_title: When dermatology residents ask a large language model to name a lesion before forming their own differential, they risk never acquiring diagnostic reasoning, which the article calls never-skilling.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: When dermatology residents ask a large language model to name a lesion before forming their own differential, they risk never acquiring diagnostic reasoning, which the article calls never-skilling.
problem_evidence: Residents increasingly ask large language models to name a lesion before diagnosing it themselves. | never-skilling, a phenomenon in which reasoning is never acquired
quick_read: The piece describes a growing habit among dermatology residents of asking large language models to identify a skin lesion before they have worked through the diagnosis on their own. It argues that this order of use creates a specific educational harm.

The harm matters because it concerns how clinical reasoning is acquired during training, not only whether a model is right or wrong. The text offers no data on how common the practice is or how much it affects skill, so the size of the risk is unclear.
limitation: The article says that forming a differential first is necessary but insufficient, so having residents reason before consulting the model does not by itself prevent the harm.
tag: Evidence-backed problem
key_points: Residents increasingly ask large language models to name a lesion before diagnosing it themselves. | The article defines never-skilling as reasoning that is never acquired in the first place. | It separates never-skilling from deskilling and from model error. | Forming a differential first is described as necessary but not sufficient. | Proposed safeguards are documented differentials, supervised disagreement, and examination of unassisted reasoning.
rundown: The article identifies a change in trainee behavior: residents are turning to large language models at the start of the diagnostic process, when they would otherwise build their own differential for a lesion.

It distinguishes this harm from two familiar concerns. Deskilling means losing an ability once held, and model error means a wrong output. Never-skilling means the reasoning was never learned at all.

For departments, the article recommends requiring documented differentials, supervising cases where resident and model disagree, and examining residents' unassisted reasoning. It says that forming a differential first is not enough by itself.
sources:
- peer_reviewed | International Journal of Dermatology | https://doi.org/10.1111/ijd.70743 | 2026-10-21
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

88ac0921710c91bf959cf51ab7eed951d36bd2c2952193388011a3c082b640f3

previous

0000000000000000000000000000000000000000000000000000000000000000