{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T10:43:59Z","timestamp":1792147439333,"version":"4.4.0"},"reference-count":19,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T00:00:00Z","timestamp":1792108800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T00:00:00Z","timestamp":1792108800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["World J Urol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study asked whether urologists can tell human-written abstracts apart from artificial intelligence (AI)-written abstracts when both describe the same study, and whether the answer depends on which AI model did the writing.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Fifteen published urology studies were selected. For each study, the original human-written abstract was paired with three AI-written abstracts covering the same material, produced by Claude Sonnet 4.6, GPT-5.5, and Gemini 3 Flash using identical instructions and each model\u2019s ordinary consumer settings. Because all four versions described the same study, the underlying data were held constant, and the versions differed in how that material was written and presented. The resulting 60 abstracts were stripped of identifying information, and eight urologists read all of them without knowing which was which (480 judgments in total). For each abstract, reviewers stated whether they believed it was written by a human or by AI, rated how confident they were in that judgment, and rated the abstract\u2019s quality and credibility on 5-point scales.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Reviewers correctly identified AI-written abstracts 49.3% of the time, approximately 50%, compared with 62.4% for human-written abstracts (\n                      <jats:italic>p<\/jats:italic>\n                      = .011). Results depended heavily on which model wrote the text (\n                      <jats:italic>p<\/jats:italic>\n                      = .003). Abstracts written by Claude were recognized as AI only 40.0% of the time, significantly less often than human abstracts were correctly identified (\n                      <jats:italic>p<\/jats:italic>\n                      = .003) and less often than GPT abstracts (56.8%,\n                      <jats:italic>p<\/jats:italic>\n                      = .038); Gemini fell between the two at 50.9%. Quality ratings differed only for GPT, which scored lower than human abstracts (\n                      <jats:italic>p<\/jats:italic>\n                      = .046; omnibus\n                      <jats:italic>p<\/jats:italic>\n                      = .030); credibility ratings did not differ between sources at all (\n                      <jats:italic>p<\/jats:italic>\n                      = .577). Reviewers who felt confident were no more likely to be correct (\n                      <jats:italic>p<\/jats:italic>\n                      = .749), and reviewers seldom agreed with one another about which abstracts were AI-written (Fleiss\u2019 \u03ba = \u22120.009).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Urologists could not reliably recognize AI-written urology abstracts, and the difficulty came mainly from one model rather than from AI writing in general. How good an abstract appeared had little to do with whether its authorship was correctly identified, and neither reviewer confidence nor agreement between reviewers gave any warning. Reading alone is therefore not a dependable way to detect undisclosed AI authorship.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00345-026-06872-6","type":"journal-article","created":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T09:46:11Z","timestamp":1792143971000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Confidence without consensus? Human versus AI abstracts in urology"],"prefix":"10.1007","volume":"44","author":[{"given":"Sholem","family":"Hack","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shlomo","family":"Resnik","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yoram","family":"Dekel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shlomi","family":"Moshe","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Noy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohammad","family":"Shalabi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ari","family":"Frajewicki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Omri","family":"Zonenfeld","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sobhi","family":"Khoury","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Israel","family":"Benzekry","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yazeed","family":"Barghouthy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-1227-3515","authenticated-orcid":false,"given":"Maharan","family":"Kabha","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,10,16]]},"reference":[{"issue":"1","key":"6872_CR1","doi-asserted-by":"publisher","first-page":"183","DOI":"10.1038\/s41746-024-01157-x","volume":"7","author":"J Haltaufderheide","year":"2024","unstructured":"Haltaufderheide J, Ranisch R (2024) The ethics of ChatGPT in medicine and healthcare: a systematic review on Large Language Models (LLMs). npj Digit Med 7(1):183. https:\/\/doi.org\/10.1038\/s41746-024-01157-x","journal-title":"npj Digit Med"},{"issue":"5","key":"6872_CR2","doi-asserted-by":"publisher","first-page":"e2401227121","DOI":"10.1073\/PNAS.2401227121","volume":"122","author":"M Binz","year":"2025","unstructured":"Binz M, Alaniz S, Roskies A et al (2025) How should the advancement of large language models affect the practice of science? Proc Natl Acad Sci U S A 122(5):e2401227121. https:\/\/doi.org\/10.1073\/PNAS.2401227121","journal-title":"Proc Natl Acad Sci U S A"},{"key":"6872_CR3","doi-asserted-by":"publisher","DOI":"10.1161\/STROKEAHA.125.051913","author":"R Khera","year":"2025","unstructured":"Khera R, Pedroso AF, Keloth VK, Xu H, Silva GS, Schwamm LH (2025) Scientific Writing in the Era of Large Language Models: A Computational Analysis of AI- Versus Human-Created Content. Stroke Published online. https:\/\/doi.org\/10.1161\/STROKEAHA.125.051913","journal-title":"Stroke Published online"},{"issue":"5","key":"6872_CR4","doi-asserted-by":"publisher","first-page":"393","DOI":"10.4196\/KJPP.2024.28.5.393","volume":"28","author":"S Ahn","year":"2024","unstructured":"Ahn S (2024) The transformative impact of large language models on medical writing and publishing: current applications, challenges and future directions. Korean J Physiol Pharmacol 28(5):393\u2013401. https:\/\/doi.org\/10.4196\/KJPP.2024.28.5.393","journal-title":"Korean J Physiol Pharmacol"},{"issue":"3","key":"6872_CR5","doi-asserted-by":"publisher","first-page":"200","DOI":"10.12982\/OSR.2025.061","volume":"46","author":"SN Asiri","year":"2025","unstructured":"Asiri SN, Alhazmi AS, Mehta V (2025) Assessing the Applicability of Artificial Intelligence (AI) in Fabrication and Plagiarism of Scientific Literature in the Field of Orthodontics. Oral Sci Rep 46(3):200\u2013210. https:\/\/doi.org\/10.12982\/OSR.2025.061","journal-title":"Oral Sci Rep"},{"issue":"1","key":"6872_CR6","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1038\/s41746-023-00819-6","volume":"6","author":"CA Gao","year":"2023","unstructured":"Gao CA, Howard FM, Markov NS et al (2023) Comparing scientific abstracts generated by ChatGPT to real abstracts with detectors and blinded human reviewers. npj Digit Med 6(1):75. https:\/\/doi.org\/10.1038\/s41746-023-00819-6","journal-title":"npj Digit Med"},{"issue":"3","key":"6872_CR7","doi-asserted-by":"publisher","first-page":"615","DOI":"10.1007\/S13304-025-02106-3","volume":"77","author":"KJ Nabata","year":"2025","unstructured":"Nabata KJ, AlShehri Y, Mashat A, Wiseman SM (2025) Evaluating human ability to distinguish between ChatGPT-generated and original scientific abstracts. Updates Surg 77(3):615\u2013621. https:\/\/doi.org\/10.1007\/S13304-025-02106-3","journal-title":"Updates Surg"},{"issue":"10","key":"6872_CR8","doi-asserted-by":"publisher","first-page":"2573","DOI":"10.1161\/STROKEAHA.124.045012","volume":"55","author":"GS Silva","year":"2024","unstructured":"Silva GS, Khera R, Schwamm LH et al (2024) Reviewer Experience Detecting and Judging Human Versus Artificial Intelligence Content: The Stroke Journal Essay Contest. Stroke 55(10):2573\u20132578. https:\/\/doi.org\/10.1161\/STROKEAHA.124.045012","journal-title":"Stroke"},{"issue":"4","key":"6872_CR9","doi-asserted-by":"publisher","first-page":"622","DOI":"10.1016\/j.mayocp.2024.08.029","volume":"100","author":"SA Helgeson","year":"2025","unstructured":"Helgeson SA, Johnson PW, Gopikrishnan N et al (2025) Human Reviewers\u2019 Ability to Differentiate Human-Authored or Artificial Intelligence-Generated Medical Manuscripts: A Randomized Survey Study. Mayo Clin Proc 100(4):622\u2013633. https:\/\/doi.org\/10.1016\/j.mayocp.2024.08.029","journal-title":"Mayo Clin Proc"},{"issue":"2","key":"6872_CR10","doi-asserted-by":"publisher","first-page":"276e","DOI":"10.1016\/j.ajog.2024.04.045","volume":"231","author":"ET Pan","year":"2024","unstructured":"Pan ET, Florian-Rodriguez M (2024) Human vs machine: identifying ChatGPT-generated abstracts in Gynecology and Urogynecology. Am J Obstet Gynecol 231(2):276e. 1-276.e10","journal-title":"Am J Obstet Gynecol"},{"issue":"1","key":"6872_CR11","doi-asserted-by":"publisher","first-page":"e52164","DOI":"10.2196\/52164","volume":"8","author":"HT Hakam","year":"2024","unstructured":"Hakam HT, Prill R, Korte L et al (2024) Human-Written vs AI-Generated Texts in Orthopedic Academic Literature: Comparative Qualitative Analysis. JMIR Form Res 8(1):e52164. https:\/\/doi.org\/10.2196\/52164","journal-title":"JMIR Form Res"},{"issue":"1","key":"6872_CR12","doi-asserted-by":"publisher","first-page":"e62779","DOI":"10.2196\/62779","volume":"11","author":"B Doru","year":"2025","unstructured":"Doru B, Maier C, Busse JS et al (2025) Detecting Artificial Intelligence-Generated Versus Human-Written Medical Student Essays: Semirandomized Controlled Study. JMIR Med Educ 11(1):e62779. https:\/\/doi.org\/10.2196\/62779","journal-title":"JMIR Med Educ"},{"key":"6872_CR13","doi-asserted-by":"publisher","unstructured":"Hack S, Attal R, Levi L et al Clarity Without Credibility? Human Versus AI Abstracts in Otolaryngology. World J Otorhinolaryngol Head Neck Surg Published online 2026:1\u20136. https:\/\/doi.org\/10.1002\/WJO2.70104","DOI":"10.1002\/WJO2.70104"},{"key":"6872_CR14","doi-asserted-by":"publisher","unstructured":"Kresoja KP, Schoeber A, Hinkel R, Papoutsis K, Thiele H (2025) Evaluation of the performance of fully artificial intelligence generated vs human-authored abstracts at a large national cardiological congress. Eur Heart J 46(Supplement1). https:\/\/doi.org\/10.1093\/EURHEARTJ\/EHAF784.4416","DOI":"10.1093\/EURHEARTJ\/EHAF784.4416"},{"issue":"1","key":"6872_CR15","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1007\/S00345-024-05146-3","volume":"42","author":"D Pompili","year":"2024","unstructured":"Pompili D, Richa Y, Collins P, Richards H, Hennessey DB (2024) Using artificial intelligence to generate medical literature for urology patients: a comparison of three different large language models. World J Urol 42(1):455. https:\/\/doi.org\/10.1007\/S00345-024-05146-3","journal-title":"World J Urol"},{"issue":"5","key":"6872_CR16","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1002\/BCO2.359","volume":"5","author":"J Eckrich","year":"2024","unstructured":"Eckrich J, Ellinger J, Cox A et al (2024) Urology consultants versus large language models: Potentials and hazards for medical advice in urology. BJUI Compass 5(5):552\u2013558. https:\/\/doi.org\/10.1002\/BCO2.359","journal-title":"BJUI Compass"},{"issue":"6","key":"6872_CR17","doi-asserted-by":"publisher","first-page":"1643","DOI":"10.1016\/j.jpurol.2025.05.030","volume":"21","author":"LA \u2019t Hoen","year":"2025","unstructured":"\u2019t Hoen LA, van Uitert A, Bussmann M et al (2025) Annual updates of the European Association of Urology - European Society for Pediatric Urology (EAU-ESPU) paediatric urology guidelines: Are large-language models (LLM) better than the usual structured methodology? J Pediatr Urol 21(6):1643\u20131649. https:\/\/doi.org\/10.1016\/j.jpurol.2025.05.030","journal-title":"J Pediatr Urol"},{"key":"6872_CR18","doi-asserted-by":"publisher","unstructured":"Hellstrom W, Singh D, Brinkley G et al (2024) (023) Can Artificial Intelligence Generate a Superior Introduction to a Systematic Review? J Sex Med. ;21(Supplement_1). https:\/\/doi.org\/10.1093\/JSXMED\/QDAE001.021","DOI":"10.1093\/JSXMED\/QDAE001.021"},{"key":"6872_CR19","first-page":"29575","volume":"235","author":"W Liang","year":"2024","unstructured":"Liang W, Izzo Z, Zhang Y et al (2024) Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews. Proc Mach Learn Res 235:29575\u201329620. https:\/\/arxiv.org\/abs\/2403.07183v4","journal-title":"Proc Mach Learn Res"}],"container-title":["World Journal of Urology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00345-026-06872-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00345-026-06872-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00345-026-06872-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T09:46:12Z","timestamp":1792143972000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00345-026-06872-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,16]]},"references-count":19,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["6872"],"URL":"https:\/\/doi.org\/10.1007\/s00345-026-06872-6","relation":{},"ISSN":["1433-8726"],"issn-type":[{"value":"1433-8726","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,16]]},"assertion":[{"value":"6 August 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 October 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 October 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no conflicts of interest.","order":1,"name":"Ethics","label":"Conflict of interest","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study used only previously published, publicly available scientific material and de-identified reviewer evaluations. No patient data or identifiable health information was accessed. Participating reviewers were practicing urology clinicians who provided evaluations in a professional capacity; no personal or sensitive data were collected from them. In accordance with prevailing regulations and institutional policies, institutional review board approval and informed consent were not required.","order":2,"name":"Ethics","label":"Ethical approval","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Artificial intelligence tools were used only for editorial and visualization-support purposes. Claude Opus 5 (Anthropic) was used to assist with grammar, clarity, and language editing during manuscript preparation, and to assist with development and formatting of figures and tables using the authors\u2019 own study data. The tool was not used for study design, data analysis, or data interpretation. All statistical analyses were performed by the authors in their local R environment, and all results, interpretations, and final manuscript content were reviewed and verified by the authors.","order":3,"name":"Ethics","label":"AI disclosure","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"751"}}