{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T07:33:49Z","timestamp":1769758429580,"version":"3.49.0"},"reference-count":10,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2026,1,29]],"date-time":"2026-01-29T00:00:00Z","timestamp":1769644800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/publication-rights-and-licensing-policy"}],"funder":[{"name":"Swiss National Science Foundation","award":["TMSGI2_211379"],"award-info":[{"award-number":["TMSGI2_211379"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Commun. ACM"],"published-print":{"date-parts":[[2026,2]]},"abstract":"<jats:p>A call to reform AI model-training paradigms from post hoc alignment to intrinsic, identity-based development.<\/jats:p>","DOI":"10.1145\/3748645","type":"journal-article","created":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T16:32:51Z","timestamp":1769617971000},"page":"24-27","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["From Model Training to Model Raising"],"prefix":"10.1145","volume":"69","author":[{"given":"Roland","family":"Aydin","sequence":"first","affiliation":[{"name":"Technische Universit\u00e4t Hamburg, Hamburg, Hamburg, Germany"},{"name":"Helmholtz-Zentrum Hereon, Geesthacht, SH, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8264-0885","authenticated-orcid":false,"given":"Christian","family":"Cyron","sequence":"additional","affiliation":[{"name":"Technische Universit\u00e4t Hamburg, Hamburg, HH, Germany"},{"name":"Helmholtz-Zentrum Hereon, Geesthacht, SH, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Steve","family":"Bachelor","sequence":"additional","affiliation":[{"name":"Independent Researcher, Stuttgart, Baden-W\u00fcrttemberg, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3089-6883","authenticated-orcid":false,"given":"Ashton","family":"Anderson","sequence":"additional","affiliation":[{"name":"University of Toronto, Toronto, Ontario, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3984-1232","authenticated-orcid":false,"given":"Robert","family":"West","sequence":"additional","affiliation":[{"name":"EPFL, Lausanne, VD, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2026,1,29]]},"reference":[{"key":"e_1_3_3_2_2","unstructured":"Betley J. et al. Emergent misalignment: Narrow finetuning can produce broadly misaligned LLMs. In Proceedings of the FM-Wild Workshop at ICLR\u00a0(2025)."},{"key":"e_1_3_3_3_2","doi-asserted-by":"crossref","unstructured":"Bui N. et al. Mixture-of-personas language models for population simulation.\u00a0Findings of the Association for Computational Linguistics: ACL 2025\u00a0(2025).","DOI":"10.18653\/v1\/2025.findings-acl.1271"},{"key":"e_1_3_3_4_2","unstructured":"Korbak T. et al. Pretraining language models with human preferences. In Proceedings of the Intern. Conf. on Machine Learning\u00a0(2023)."},{"key":"e_1_3_3_5_2","unstructured":"Maini \u00a0 P . et al. Safety pretraining: Toward the next generation of safe AI. arXiv (April 2025)."},{"key":"e_1_3_3_6_2","unstructured":"Mehrotra A. et al. Tree of attacks: Jailbreaking black-box LLMs automatically.\u00a0Advances in Neural Information Processing Systems\u00a0(2024)."},{"key":"e_1_3_3_7_2","unstructured":"Qi X. et al. Fine-tuning aligned language models compromises safety even when users do not intend to!\u00a0In Proceedings of the Intern. Conf. on Learning Representations\u00a0(2024)."},{"key":"e_1_3_3_8_2","volume-title":"Human Compatible: Artificial Intelligence and the Problem of Control","author":"Russell S.","year":"2019","unstructured":"Russell, S. Human Compatible: Artificial Intelligence and the Problem of Control. Viking, New York\u00a0(2019)."},{"key":"e_1_3_3_9_2","doi-asserted-by":"publisher","DOI":"10.1145\/3705294"},{"key":"e_1_3_3_10_2","unstructured":"Wolf Y. et al. Fundamental limitations of alignment in large language models. In Proceedings of the Intern. Conf. on Machine Learning\u00a0(2024)."},{"key":"e_1_3_3_11_2","unstructured":"Zou A. et al. Universal and transferable adversarial attacks on aligned language models. arXiv:2307.15043 (2023)."}],"container-title":["Communications of the ACM"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3748645","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,29]],"date-time":"2026-01-29T17:05:25Z","timestamp":1769706325000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3748645"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,29]]},"references-count":10,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,2]]}},"alternative-id":["10.1145\/3748645"],"URL":"https:\/\/doi.org\/10.1145\/3748645","relation":{},"ISSN":["0001-0782","1557-7317"],"issn-type":[{"value":"0001-0782","type":"print"},{"value":"1557-7317","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,29]]},"assertion":[{"value":"2025-06-22","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-29","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}