{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T22:32:38Z","timestamp":1793140358013,"version":"4.5.0"},"reference-count":35,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T00:00:00Z","timestamp":1788220800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T00:00:00Z","timestamp":1786665600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["clinicalkey.com","clinicalkey.com.au","clinicalkey.es","clinicalkey.fr","clinicalkey.jp","thelancet.com","elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["eClinicalMedicine"],"published-print":{"date-parts":[[2026,9]]},"DOI":"10.1016\/j.eclinm.2026.104184","type":"journal-article","created":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T15:04:19Z","timestamp":1788188659000},"page":"104184","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["The benchmark illusion: on the curious underestimation of AI and the uncertainty of its creators\u2019 gold standards"],"prefix":"10.1016","volume":"99","author":[{"given":"Christian","family":"Pohlkamp","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Torsten","family":"Haferlach","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"issue":"7972","key":"10.1016\/j.eclinm.2026.104184_bib1","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","article-title":"Large language models encode clinical knowledge","volume":"620","author":"Singhal","year":"2023","journal-title":"Nature"},{"issue":"3","key":"10.1016\/j.eclinm.2026.104184_bib2","doi-asserted-by":"crossref","first-page":"943","DOI":"10.1038\/s41591-024-03423-7","article-title":"Toward expert-level medical question answering with large language models","volume":"31","author":"Singhal","year":"2025","journal-title":"Nat Med"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib3","doi-asserted-by":"crossref","DOI":"10.1080\/10872981.2023.2220920","article-title":"ChatGPT in medical school: how successful is AI in progress testing?","volume":"28","author":"Friederichs","year":"2023","journal-title":"Med Educ Online"},{"issue":"8067","key":"10.1016\/j.eclinm.2026.104184_bib5","doi-asserted-by":"crossref","first-page":"442","DOI":"10.1038\/s41586-025-08866-7","article-title":"Towards conversational diagnostic artificial intelligence","volume":"642","author":"Tu","year":"2025","journal-title":"Nature"},{"issue":"8","key":"10.1016\/j.eclinm.2026.104184_bib6","doi-asserted-by":"crossref","first-page":"1836","DOI":"10.1093\/annonc\/mdy166","article-title":"Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists","volume":"29","author":"Haenssle","year":"2018","journal-title":"Ann Oncol"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib7","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1038\/s41746-024-01103-x","article-title":"A systematic review and meta-analysis of artificial intelligence versus clinicians for skin cancer diagnosis","volume":"7","author":"Salinas","year":"2024","journal-title":"NPJ Digit Med"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib8","doi-asserted-by":"crossref","first-page":"114","DOI":"10.1038\/s41746-024-01106-8","article-title":"Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy","volume":"7","author":"McGenity","year":"2024","journal-title":"NPJ Digit Med"},{"issue":"22","key":"10.1016\/j.eclinm.2026.104184_bib9","doi-asserted-by":"crossref","first-page":"2199","DOI":"10.1001\/jama.2017.14585","article-title":"Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer","volume":"318","author":"Ehteshami Bejnordi","year":"2017","journal-title":"JAMA"},{"issue":"7788","key":"10.1016\/j.eclinm.2026.104184_bib10","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1038\/s41586-019-1799-6","article-title":"International evaluation of an AI system for breast cancer screening","volume":"577","author":"McKinney","year":"2020","journal-title":"Nature"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib11","doi-asserted-by":"crossref","first-page":"116","DOI":"10.1016\/S1470-2045(25)00567-4","article-title":"Artificial intelligence and radiologists in pancreatic cancer detection using standard of care CT scans (PANORAMA): an international, paired, non-inferiority, confirmatory, observational study","volume":"27","author":"Alves","year":"2026","journal-title":"Lancet Oncol"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib12","doi-asserted-by":"crossref","first-page":"1209","DOI":"10.1186\/s12909-025-07761-z","article-title":"AI-powered platform revolutionizing blood cell morphology education for medical students","volume":"25","author":"Liu","year":"2025","journal-title":"BMC Med Educ"},{"issue":"11","key":"10.1016\/j.eclinm.2026.104184_bib13","doi-asserted-by":"crossref","first-page":"1791","DOI":"10.1038\/s42256-025-01122-7","article-title":"Deep generative classification of blood cell morphology","volume":"7","author":"Deltadahl","year":"2025","journal-title":"Nat Mach Intell"},{"issue":"11","key":"10.1016\/j.eclinm.2026.104184_bib14","doi-asserted-by":"crossref","first-page":"1571","DOI":"10.1016\/j.mayocp.2018.05.014","article-title":"Physician burnout, well-being, and work unit safety grades in relationship to reported medical errors","volume":"93","author":"Tawfik","year":"2018","journal-title":"Mayo Clin Proc"},{"issue":"12","key":"10.1016\/j.eclinm.2026.104184_bib15","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2020.28111","article-title":"Assessment of physician sleep and wellness, burnout, and clinically significant medical errors","volume":"3","author":"Trockel","year":"2020","journal-title":"JAMA Netw Open"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib16","article-title":"Should artificial intelligence have lower acceptable error rates than humans?","volume":"5","author":"Lenskjold","year":"2023","journal-title":"BJR Open"},{"key":"10.1016\/j.eclinm.2026.104184_bib17","article-title":"Prevalence, severity, and nature of preventable patient harm across medical care settings: systematic review and meta-analysis","volume":"366","author":"Panagioti","year":"2019","journal-title":"BMJ"},{"issue":"2","key":"10.1016\/j.eclinm.2026.104184_bib18","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1001\/jamainternmed.2023.7347","article-title":"Diagnostic errors in hospitalized adults who died or were transferred to intensive care","volume":"184","author":"Auerbach","year":"2024","journal-title":"JAMA Intern Med"},{"issue":"12","key":"10.1016\/j.eclinm.2026.104184_bib19","doi-asserted-by":"crossref","first-page":"1008","DOI":"10.1136\/bmjqs-2019-010822","article-title":"Prevalence of harmful diagnostic errors in hospitalised adults: a systematic review and meta-analysis","volume":"29","author":"Gunderson","year":"2020","journal-title":"BMJ Qual Saf"},{"issue":"9","key":"10.1016\/j.eclinm.2026.104184_bib20","doi-asserted-by":"crossref","first-page":"727","DOI":"10.1136\/bmjqs-2013-002627","article-title":"The frequency of diagnostic errors in outpatient care: estimations from three large observational studies involving US adult populations","volume":"23","author":"Singh","year":"2014","journal-title":"BMJ Qual Saf"},{"key":"10.1016\/j.eclinm.2026.104184_bib21","series-title":"Improving Diagnosis in Health Care","year":"2015"},{"issue":"11","key":"10.1016\/j.eclinm.2026.104184_bib22","doi-asserted-by":"crossref","first-page":"1184","DOI":"10.1197\/aemj.9.11.1184","article-title":"Achieving quality in clinical decision making: cognitive strategies and detection of bias","volume":"9","author":"Croskerry","year":"2002","journal-title":"Acad Emerg Med"},{"issue":"Suppl 2","key":"10.1016\/j.eclinm.2026.104184_bib23","doi-asserted-by":"crossref","first-page":"ii58","DOI":"10.1136\/bmjqs-2012-001712","article-title":"Cognitive debiasing 1: origins of bias and theory of debiasing","volume":"22","author":"Croskerry","year":"2013","journal-title":"BMJ Qual Saf"},{"issue":"13","key":"10.1016\/j.eclinm.2026.104184_bib24","doi-asserted-by":"crossref","first-page":"1493","DOI":"10.1001\/archinte.165.13.1493","article-title":"Diagnostic error in internal medicine","volume":"165","author":"Graber","year":"2005","journal-title":"Arch Intern Med"},{"issue":"3","key":"10.1016\/j.eclinm.2026.104184_bib25","doi-asserted-by":"crossref","first-page":"337","DOI":"10.14744\/nci.2025.11456","article-title":"Assessment of interobserver variability in Gleason grading for prostate carcinoma","volume":"12","author":"Cinar","year":"2025","journal-title":"North Clin Istanb"},{"issue":"3","key":"10.1016\/j.eclinm.2026.104184_bib26","doi-asserted-by":"crossref","first-page":"360","DOI":"10.3324\/haematol.2011.047811","article-title":"European bone marrow working group trial on reproducibility of world health organization criteria to discriminate essential thrombocythemia from prefibrotic primary myelofibrosis","volume":"97","author":"Buhr","year":"2012","journal-title":"Haematologica"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib27","doi-asserted-by":"crossref","first-page":"175","DOI":"10.1038\/s41746-025-01543-z","article-title":"A systematic review and meta-analysis of diagnostic performance comparison between generative AI and physicians","volume":"8","author":"Takita","year":"2025","journal-title":"NPJ Digit Med"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib28","doi-asserted-by":"crossref","first-page":"190","DOI":"10.1038\/s41746-024-01185-7","article-title":"Hidden flaws behind expert-level accuracy of multimodal GPT-4 vision in medicine","volume":"7","author":"Jin","year":"2024","journal-title":"NPJ Digit Med"},{"issue":"1","key":"10.1016\/j.eclinm.2026.104184_bib29","doi-asserted-by":"crossref","first-page":"31","DOI":"10.1038\/s41746-021-00385-9","article-title":"Do as AI say: susceptibility in deployment of clinical decision-aids","volume":"4","author":"Gaube","year":"2021","journal-title":"NPJ Digit Med"},{"issue":"11","key":"10.1016\/j.eclinm.2026.104184_bib30","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pdig.0000651","article-title":"Bias in medical AI: implications for clinical decision-making","volume":"3","author":"Cross","year":"2024","journal-title":"PLOS Digit Health"},{"issue":"8","key":"10.1016\/j.eclinm.2026.104184_bib31","doi-asserted-by":"crossref","first-page":"775","DOI":"10.1097\/00001888-200308000-00003","article-title":"The importance of cognitive errors in diagnosis and strategies to minimize them","volume":"78","author":"Croskerry","year":"2003","journal-title":"Acad Med"},{"issue":"26","key":"10.1016\/j.eclinm.2026.104184_bib32","doi-asserted-by":"crossref","first-page":"2445","DOI":"10.1056\/NEJMp1303712","article-title":"From mindless to mindful practice--cognitive bias and clinical decision making","volume":"368","author":"Croskerry","year":"2013","journal-title":"N Engl J Med"},{"key":"10.1016\/j.eclinm.2026.104184_bib33","series-title":"FG-AI4H DEL7.4: Clinical Evaluation of AI for Health","author":"Union","year":"2023"},{"key":"10.1016\/j.eclinm.2026.104184_bib35","doi-asserted-by":"crossref","DOI":"10.1136\/bmj-2024-081554","article-title":"FUTURE-AI: international consensus guideline for trustworthy and deployable artificial intelligence in healthcare","volume":"388","author":"Lekadir","year":"2025","journal-title":"Bmj"},{"key":"10.1016\/j.eclinm.2026.104184_bib36","article-title":"Evaluating AI systems under uncertain ground truth: a case study in dermatology","author":"Stutz","year":"2023","journal-title":"arXiv"},{"issue":"8","key":"10.1016\/j.eclinm.2026.104184_bib37","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1001\/jama.2024.0549","article-title":"Regulate artificial intelligence in health care by prioritizing patient outcomes","volume":"331","author":"Ayers","year":"2024","journal-title":"JAMA"}],"container-title":["eClinicalMedicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S2589537026004372?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S2589537026004372?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,10,27]],"date-time":"2026-10-27T21:34:22Z","timestamp":1793136862000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S2589537026004372"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9]]},"references-count":35,"alternative-id":["S2589537026004372"],"URL":"https:\/\/doi.org\/10.1016\/j.eclinm.2026.104184","relation":{},"ISSN":["2589-5370"],"issn-type":[{"value":"2589-5370","type":"print"}],"subject":[],"published":{"date-parts":[[2026,9]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"The benchmark illusion: on the curious underestimation of AI and the uncertainty of its creators\u2019 gold standards","name":"articletitle","label":"Article Title"},{"value":"eClinicalMedicine","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.eclinm.2026.104184","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 The Author(s). Published by Elsevier Ltd.","name":"copyright","label":"Copyright"}],"article-number":"104184"}}