{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T11:09:07Z","timestamp":1788952147075,"version":"build-2803163510"},"reference-count":13,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T00:00:00Z","timestamp":1788912000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    Modern clinical epidemiology and artificial intelligence are increasingly driven by an idealized premise: the belief that massive databases and advanced machine learning can redeem causal inference from observational uncertainty. Behind the facade of multi-million-record cohorts, however, a profound epistemological crisis festers, revealing a potentially sick artificial intelligence. Modern analytical systems leverage massive sample sizes to generate ultra-significant asymptotic\n                    <jats:italic>p<\/jats:italic>\n                    -values that reflect mathematical outcomes rather than biological reality. This article diagnoses this systemic pathology, the uncritical application of Gaussian asymptotic statistics to sparse, discrete, and rare medical event counts governed by Poisson distributions. To cure this condition, we introduce a methodological antidote: an inference stress test that repurposes the exact lower boundary of the exact Poisson confidence interval as a formal null benchmark, utilizing an event-anchored standard error. Across three clinical case studies, spanning autoimmune dermatology baselines, expanded matching cohorts, and oncological lifestyle investigations, our stress test provides evidence that nominal asymptotic significance may sometimes fail to clear the structural resistance threshold, revealing the underlying fragility of scale-induced signals. This proposed framework serves as an essential epistemological filter, separating genuine biological discovery from digital outcomes and ensuring that future medical AI systems learn only from structurally validated knowledge.\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1961525","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T10:12:10Z","timestamp":1788948730000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["An AI for diseases or a sick AI? The epistemological pathology of big data overload and the loss of statistical significance"],"prefix":"10.3389","volume":"9","author":[{"given":"Marco","family":"Roccetti","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering Bologna, University of Bologna","place":["Bologna, Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,9]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1080\/00031305.1998.10480550","article-title":"Approximate is better than exact for interval estimation of binomial proportions","volume":"52","author":"Agresti","year":"1998","journal-title":"Am. Stat."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"e68898","DOI":"10.2196\/68898","article-title":"The use of machine learning for analyzing real-world data in disease prediction and management: systematic review","volume":"13","author":"Alhumaidi","year":"2025","journal-title":"JMIR Med. Inform."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"e061602","DOI":"10.1136\/bmjopen-2022-061602","article-title":"Questioning the seasonality of SARS-COV-2: a Fourier spectral analysis","volume":"12","author":"Cappi","year":"2022","journal-title":"BMJ Open"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"794","DOI":"10.3390\/e26090794","article-title":"Is seeing believing? A practitioner\u2019s perspective on high-dimensional statistical inference in Cancer genomics studies","volume":"26","author":"Fan","year":"2024","journal-title":"Entropy"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"437","DOI":"10.1093\/biomet\/28.3-4.437","article-title":"Fiducial limits for the Poisson distribution","volume":"28","author":"Garwood","year":"1936","journal-title":"Biometrika"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"291","DOI":"10.2174\/1389202922666210705124359","article-title":"Current genomics, machine learning in healthcare","volume":"22","author":"Habehh","year":"2021","journal-title":"Curr. Genomics"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"614","DOI":"10.1016\/j.jaci.2026.06.009","article-title":"Reply","volume":"158","author":"Kim","year":"2026","journal-title":"J. Allergy Clin. Immunol."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1016\/j.jaci.2024.01.025","article-title":"Autoimmune adverse event following COVID-19 vaccination in Seoul, South Korea","volume":"153","author":"Kim","year":"2025","journal-title":"J. Allergy Clin. Immunol."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"101752","DOI":"10.1016\/j.imu.2026.101752","article-title":"Unlocking the stochastic parrot: epistemic obligation and the decline of biological plausibility in clinical reality","volume":"62","author":"Roccetti","year":"","journal-title":"Inform. Med. Unlocked."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"614","DOI":"10.1016\/j.jaci.2026.02.023","article-title":"Logical paradoxes and epidemiologic inconsistencies in vaccine-related vitiligo risk assessment","volume":"158","author":"Roccetti","year":"","journal-title":"J. Allergy Clin. Immunol."},{"key":"ref11","volume-title":"Modern Epidemiology","author":"Rothman","year":"2008"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"116869","DOI":"10.1016\/j.ejca.2026.116869","article-title":"Association between lifestyle and risk of early-onset cancer: evidence from the European prospective investigation into cancer and nutrition and the UK biobank","volume":"243","author":"Sala","year":"2026","journal-title":"Eur. J. Cancer"},{"key":"ref13","author":"Therneau","year":"1997"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1961525\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T10:12:11Z","timestamp":1788948731000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1961525\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,9]]},"references-count":13,"alternative-id":["10.3389\/frai.2026.1961525"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1961525","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,9]]},"article-number":"1961525"}}