{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T11:51:44Z","timestamp":1791978704986,"version":"4.4.0"},"reference-count":15,"publisher":"eLife Sciences Publications, Ltd","license":[{"start":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T00:00:00Z","timestamp":1791936000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T00:00:00Z","timestamp":1791936000000},"content-version":"am","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T00:00:00Z","timestamp":1791936000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["elifesciences.org"],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Large language models are very good at providing concise answers. But if we only see the answer without knowing the original question or the context behind it, that answer may not be very useful. The important issue is not simply how much information an AI summary removes but whether we can still recover the missing context. Thus, AI literature tools should also report how much information was removed and how recoverable their omissions are.<\/jats:p>","DOI":"10.7554\/elife.113137","type":"journal-article","created":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T15:15:07Z","timestamp":1791472507000},"update-policy":"https:\/\/doi.org\/10.7554\/elife.113137","source":"Crossref","is-referenced-by-count":0,"title":["The 42 problem in AI-assisted research"],"prefix":"10.7554","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6812-0979","authenticated-orcid":true,"given":"Giovanni","family":"Spitale","sequence":"first","affiliation":[{"id":[{"id":"https:\/\/ror.org\/02crff812","id-type":"ROR","asserted-by":"publisher"}],"name":"ITE Lab, Institute of Biomedical Ethics and History of Medicine, University of Zurich","place":["Zurich, Switzerland"]}],"role":[{"role":"author","vocabulary":"crossref"},{"role":"corresponding-author","vocabulary":"crossref"},{"role":"conceptualization","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"project-administration","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5604-0437","authenticated-orcid":true,"given":"Federico","family":"Germani","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/02crff812","id-type":"ROR","asserted-by":"publisher"}],"name":"ITE Lab, Institute of Biomedical Ethics and History of Medicine, University of Zurich","place":["Zurich, Switzerland"]},{"id":[{"id":"https:\/\/ror.org\/03z9tma90","id-type":"ROR","asserted-by":"publisher"}],"name":"Institute for Data Science and Artificial Intelligence, Bo\u011fazi\u00e7i University","place":["Istanbul, Turkiye"]}],"role":[{"role":"author","vocabulary":"crossref"},{"role":"conceptualization","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]}],"member":"4374","published-online":{"date-parts":[[2026,10,14]]},"reference":[{"key":"bib1","doi-asserted-by":"publisher","DOI":"10.1186\/s13643-026-03199-6","type":"journal-article","article-title":"Comparing supervised machine learning and large language models in title-abstract screening","volume":"15","author":"Aigner","year":"2026","journal-title":"Systematic Reviews"},{"key":"bib2","doi-asserted-by":"publisher","first-page":"212","DOI":"10.5334\/pme.2431","type":"journal-article","article-title":"The presence and nature of AI-use disclosure statements in medical education journals: a bibliometric study","volume":"15","author":"Ans","year":"2026","journal-title":"Perspectives on Medical Education"},{"key":"bib3","doi-asserted-by":"publisher","DOI":"10.1057\/s41599-021-00903-w","type":"journal-article","article-title":"Growth rates of modern science: a latent piecewise growth curve approach to model publication numbers from established and new literature databases","volume":"8","author":"Bornmann","year":"2021","journal-title":"Humanities and Social Sciences Communications"},{"key":"bib4","type":"web-resource","article-title":"Authorship and AI tools","author":"COPE","year":"2023","unstructured":"COPE. 2023. Authorship and AI tools. https:\/\/publicationethics.org\/guidance\/cope-position\/authorship-and-ai-tools [Accessed August 27, 2026]."},{"key":"bib5","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adz2924","type":"journal-article","article-title":"Source framing triggers systematic bias in large language models","volume":"11","author":"Germani","year":"2025","journal-title":"Science Advances"},{"key":"bib6","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.b2680","type":"journal-article","article-title":"How citation distortions create unfounded authority: analysis of a citation network","volume":"339","author":"Greenberg","year":"2009","journal-title":"BMJ"},{"key":"bib7","type":"web-resource","article-title":"Recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals","author":"ICMJE","year":"2026","unstructured":"ICMJE. 2026. Recommendations for the conduct, reporting, editing, and publication of scholarly work in medical journals. https:\/\/www.icmje.org\/recommendations\/ [Accessed August 27, 2026]."},{"key":"bib8","type":"web-resource","article-title":"eLife Latest: Exploring AI\u2019s potential for improving research engagement","author":"Inside eLife","year":"2026","unstructured":"Inside eLife. 2026. eLife Latest: Exploring AI\u2019s potential for improving research engagement. https:\/\/elifesciences.org\/inside-elife\/6ad10e4e\/elife-latest-exploring-ai-s-potential-for-improving-research-engagement [Accessed September 30, 2026]."},{"key":"bib9","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2605.06635","type":"preprint","article-title":"Cited but not verified: parsing and evaluating source attribution in LLM deep research agents","volume-title":"arXiv","author":"Onweller","year":"2026"},{"key":"bib10","doi-asserted-by":"publisher","DOI":"10.1098\/rsos.241776","type":"journal-article","article-title":"Generalization bias in large language model summarization of scientific research","volume":"12","author":"Peters","year":"2025","journal-title":"Royal Society Open Science"},{"key":"bib11","doi-asserted-by":"publisher","first-page":"30956","DOI":"10.18653\/v1\/2026.acl-long.1430","type":"conference-paper","article-title":"Attribution, citation, and quotation: A survey of evidence-based text generation with large language models","author":"Schreieder","year":"2026","unstructured":"Schreieder T, Schopf T, F\u00e4rber M. 2026. Attribution, citation, and quotation: A survey of evidence-based text generation with large language models. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)."},{"key":"bib12","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2409.13740","type":"preprint","article-title":"Language agents achieve superhuman synthesis of scientific knowledge","volume-title":"arXiv","author":"Skarlinski","year":"2024"},{"key":"bib13","doi-asserted-by":"publisher","DOI":"10.1016\/j.heliyon.2020.e04426","type":"journal-article","article-title":"Making sense in the flood. How to cope with the massive flow of digital information in medical ethics","volume":"6","author":"Spitale","year":"2020","journal-title":"Heliyon"},{"key":"bib14","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adh1850","type":"journal-article","article-title":"AI model GPT-3 (dis)informs us better than humans","volume":"9","author":"Spitale","year":"2023","journal-title":"Science Advances"},{"key":"bib15","doi-asserted-by":"publisher","first-page":"1134","DOI":"10.1038\/s41591-024-02855-5","type":"journal-article","article-title":"Adapted large language models can outperform medical experts in clinical text summarization","volume":"30","author":"Van Veen","year":"2024","journal-title":"Nature Medicine"}],"container-title":["eLife"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/cdn.elifesciences.org\/articles\/113137\/elife-113137-v1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/cdn.elifesciences.org\/articles\/113137\/elife-113137-v1.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/elifesciences.org\/articles\/113137","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T11:25:45Z","timestamp":1791977145000},"score":1,"resource":{"primary":{"URL":"https:\/\/elifesciences.org\/articles\/113137"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,14]]},"references-count":15,"alternative-id":["10.7554\/eLife.113137"],"URL":"https:\/\/doi.org\/10.7554\/elife.113137","archive":["CLOCKSS"],"relation":{},"ISSN":["2050-084X"],"issn-type":[{"value":"2050-084X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,14]]},"article-number":"e113137"}}