<HashMap><database>biostudies-literature</database><scores/><additional><omics_type>Unknown</omics_type><volume>5(6)</volume><submitter>Knochel TD</submitter><pubmed_abstract>In a rapidly evolving Generative Artificial Intelligence (GenAI) landscape, researchers, policymakers, and publishers have to continuously redefine responsible research practices. To ensure guidance of GenAI use in research, core principles that remain stable despite technological advancement are needed. This article defines a list of principles guiding the responsible use of GenAI in research, regardless of use case and GenAI technology employed. To define this framework, we conducted an anonymised Delphi consensus procedure comprising a panel of 16 international and multidisciplinary experts in AI, social sciences, law, ethics, and scientific publishing. After three rounds of independent rating and feedback, the panel reached consensus on eight sequentially ordered principles required fo</pubmed_abstract><journal>AI and ethics</journal><pagination>6371-6377</pagination><full_dataset_link>https://www.ebi.ac.uk/biostudies/studies/S-EPMC12592249</full_dataset_link><repository>biostudies-literature</repository><pubmed_title>Core principles of responsible generative AI usage in research.</pubmed_title><pmcid>PMC12592249</pmcid><pubmed_authors>Kovacs M</pubmed_authors><pubmed_authors>Yang YF</pubmed_authors><pubmed_authors>Liu M</pubmed_authors><pubmed_authors>Peeters A</pubmed_authors><pubmed_authors>Knochel TD</pubmed_authors><pubmed_authors>Heyselaar E</pubmed_authors><pubmed_authors>Aczel B</pubmed_authors><pubmed_authors>Lin Z</pubmed_authors><pubmed_authors>Al-Hoorie AH</pubmed_authors><pubmed_authors>Acar OA</pubmed_authors><pubmed_authors>Hosseini M</pubmed_authors><pubmed_authors>Vranka MA</pubmed_authors><pubmed_authors>Buehler F</pubmed_authors><pubmed_authors>Elsherif MM</pubmed_authors><pubmed_authors>Giannini A</pubmed_authors><pubmed_authors>Yamada Y</pubmed_authors><pubmed_authors>Akil AM</pubmed_authors><pubmed_authors>Ilangovan V</pubmed_authors><pubmed_authors>Ravenzwaaij DV</pubmed_authors><pubmed_authors>Schweizer KJ</pubmed_authors></additional><is_claimable>false</is_claimable><name>Core principles of responsible generative AI usage in research.</name><description>In a rapidly evolving Generative Artificial Intelligence (GenAI) landscape, researchers, policymakers, and publishers have to continuously redefine responsible research practices. To ensure guidance of GenAI use in research, core principles that remain stable despite technological advancement are needed. This article defines a list of principles guiding the responsible use of GenAI in research, regardless of use case and GenAI technology employed. To define this framework, we conducted an anonymised Delphi consensus procedure comprising a panel of 16 international and multidisciplinary experts in AI, social sciences, law, ethics, and scientific publishing. After three rounds of independent rating and feedback, the panel reached consensus on eight sequentially ordered principles required fo</description><dates><release>2025-01-01T00:00:00Z</release><publication>2025</publication><modification>2026-06-05T14:46:25.955Z</modification><creation>2026-05-18T03:08:34.856Z</creation></dates><accession>S-EPMC12592249</accession><cross_references><pubmed>41209329</pubmed><doi>10.1007/s43681-025-00768-8</doi></cross_references></HashMap>