When expertise fails: detection of artificial intelligence - generated vitiligo imagery across the spectrum of dermatological experience; a cross-sectional exploratory study
6 affiliations
- 1Maccabi Healthcare Services, Tel Aviv, 6812509 Israel
- 2Rutgers School of Public Health, Rutgers University, Piscataway, NJ 08901 USA
- 3Ono Academic College, Kiryat Ono, 55000 Israel
- 4Gray Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, 6997801 Israel
- 5Dermatology Department, Sheba Medical Center, Ramat-Gan, 5262000 Israel
- 6Division of Dermatology, Rabin Medical Center, Beilinson Hospital, Petach Tikva, 4941492 Israel
Abstract
Generative artificial intelligence (AI) can produce highly realistic dermatological images, raising concerns about clinicians’ ability to distinguish authentic pathology from synthetic representations. To assess whether dermatological expertise influences the detection of AI-generated vitiligo images, and to evaluate the self-detection capabilities of the algorithm, we conducted a participant-blinded cross-sectional study in which 31 dermatologists, stratified by experience, classified one AI-generated vitiligo image alongside three real controls while blinded to the number and identity of the synthetic images. Interrater agreement was assessed using Fleiss’ kappa statistic, and the synthetic image was additionally evaluated using three large multimodal models. Residents detected the AI-generated image more frequently than senior dermatologists and key opinion leaders. After disclosure of the correct classifications, the experts identified specific biological inconsistencies, whereas residents relied primarily on intuitive impressions, and the performance of the AI models varied across the detection and post-disclosure artifact-recognition tasks. This study is limited by its evaluation of a single AI-generated image depicting a single disease (vitiligo), and the findings may therefore not be generalizable to other dermatological morphologies, image-generation prompts, or multi-image scenarios. In this exploratory study, greater clinical experience was associated with reduced detection of the evaluated synthetic image; whether this reflects a general effect of expertise requires further study. These findings highlight the emerging risks associated with medical AI and support the integration of AI-focused visual analysis into dermatology education.
Related
- Why Generative Artificial Intelligence Makes Biomedical Expertise More, Not Less, Important
- Artificial intelligence-enabled healthcare access and user experience: comparative institutional analysis and survey evidence from China
- Effectiveness and user experience of ambient artificial intelligence-driven scribe technology among clinicians: a protocol for a systematic review.
- Learning and deskilling effects of artificial intelligence in colonoscopy among endoscopists with different levels of experience: a pragmatic, prospective trial
- Artificial intelligence-assisted smartphone thermal imaging for pedicled facial and head-neck flaps: An exploratory feasibility study using ΔT-based proxy perfusion grading.
BibTeX
@article{shapiro2026when,
title = {{When expertise fails: detection of artificial intelligence - generated vitiligo imagery across the spectrum of dermatological experience; a cross-sectional exploratory study}},
author = {Jonathan Shapiro and Itay Cohen and Mor Atlas and Anna Lyakhovitsky and Shany Sherman and Riad Kassem},
journal = {Visual Computing for Industry, Biomedicine, and Art},
year = {2026},
month = nov,
doi = {10.1186/s42492-026-00239-9},
url = {https://doi.org/10.1186/s42492-026-00239-9},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - When expertise fails: detection of artificial intelligence - generated vitiligo imagery across the spectrum of dermatological experience; a cross-sectional exploratory study
AU - Jonathan Shapiro
AU - Itay Cohen
AU - Mor Atlas
AU - Anna Lyakhovitsky
AU - Shany Sherman
AU - Riad Kassem
PY - 2026
DA - 2026/11/06
JO - Visual Computing for Industry, Biomedicine, and Art
DO - 10.1186/s42492-026-00239-9
UR - https://doi.org/10.1186/s42492-026-00239-9
AB - Generative artificial intelligence (AI) can produce highly realistic dermatological images, raising concerns about clinicians’ ability to distinguish authentic pathology from synthetic representations. To assess whether dermatological expertise influences the detection of AI-generated vitiligo images, and to evaluate the self-detection capabilities of the algorithm, we conducted a participant-blinded cross-sectional study in which 31 dermatologists, stratified by experience, classified one AI-generated vitiligo image alongside three real controls while blinded to the number and identity of the synthetic images. Interrater agreement was assessed using Fleiss’ kappa statistic, and the synthetic image was additionally evaluated using three large multimodal models. Residents detected the AI-generated image more frequently than senior dermatologists and key opinion leaders. After disclosure of the correct classifications, the experts identified specific biological inconsistencies, whereas residents relied primarily on intuitive impressions, and the performance of the AI models varied across the detection and post-disclosure artifact-recognition tasks. This study is limited by its evaluation of a single AI-generated image depicting a single disease (vitiligo), and the findings may therefore not be generalizable to other dermatological morphologies, image-generation prompts, or multi-image scenarios. In this exploratory study, greater clinical experience was associated with reduced detection of the evaluated synthetic image; whether this reflects a general effect of expertise requires further study. These findings highlight the emerging risks associated with medical AI and support the integration of AI-focused visual analysis into dermatology education.
SN - 2096-496X
SN - 2524-4442
ER -