From generative AI to epistemic collapse: Epistemically hostile environments in high-risk defense contexts
Entornos epistémicamente hostiles en contextos de defensa de alto riesgo
DOI:
https://doi.org/10.30973/esdi.2026.12.1.259Keywords:
artificial intelligence, human in the loop, meaningful human oversight, facts and values, local epistemic collapseAbstract
This article examines how artificial intelligence (AI) systems used in high-risk defense contexts may contribute to the emergence of epistemically hostile environments. Drawing on a conceptual reconstruction of different forms of epistemic mediation associated with AI—rule-based systems, machine learning, and generative AI—it analyzes how these technologies reshape the relationship between information, human judgment, and the articulation of facts and values in decision-making under uncertainty. The article argues that the mere presence of a human in the loop does not by itself guarantee control, accountability, or legitimacy, since human oversight is meaningful only when it preserves effective conditions for understanding, critical review, disagreement, and veto. On this basis, it introduces the notion of local epistemic collapse to describe situations in which algorithmic mediation degrades the human capacity to interpret contexts, question recommendations, and assume responsibility for decisions. The article concludes that responsible AI governance in defense should be assessed not only in terms of accuracy or efficiency, but also in terms of its impact on the epistemic quality of the sociotechnical environment in which judgment is exercised.
Downloads
References
ALPAC. Language and machines: computers in translation and linguistics: a report by the automatic language processing advisory committee, division of behavioral sciences, National Academy of Sciences, National Research Council. National Academy of Sciences, 1966.
Balistier, Thomas. The phaistos disk: an account of its unresolved mystery. Verlag, 2000.
Battistelli, Fabrizio y Maria Grazia Galantino. “Dangers, risks and threats: An alternative conceptualization to the catch-all concept of risk”. Current Sociology, vol. 67, núm. 1, 2019, pp. 64–78. https://doi.org/10.1177/0011392118793675
Bommasani, Rishi, et al. “On the opportunities and risks of foundation models”. arXiv, 2022. https://doi.org/10.48550/arXiv.2108.07258
Bruun, Laura y Mata Bo. Bias in Military Artificial Intelligence and Compliance with International Humanitarian Law. Stockholm International Peace Research Institute, 2025. https://doi.org/10.55163/NLWV5347
Claverie, Bernard y François Du Cluzel. “‘Cognitive warfare’: The advent of the concept of ‘cognitics’ in the field of warfare”. Cognitive warfare: the future of cognitive dominance, editado por B. Claverie, B. Prébot, N. Buchler y F. Du Cluzel. NATO Science and Technology Organization, 2022.
Cranor, Lorrie Faith. “A framework for reasoning about the human in the loop”. Proceedings of the 1st Conference on usability, psychology, and security. USENIX Association, 2008, pp. 1-15.
Deppe, Christoph y Gary Schaal. “Cognitive warfare: A conceptual analysis of the NATO ACT cognitive warfare exploratory concept”. Frontiers in Big Data, vol. 7, 2024, article 1452129. https://doi.org/10.3389/fdata.2024.1452129
Ebert, Philip, et al. “Varieties of risk”. Philosophy and Phenomenological Research, vol. 101, núm. 2, 2020, pp. 432–455. https://doi.org/10.1111/phpr.12598
Farnós, Joan, et al. “Ethical prompting: Toward strategies for rapid and inclusive assistance in dual-use AI systems”. Frontiers in Artificial Intelligence, vol. 8, 2025, 1646444. https://doi.org/10.3389/frai.2025.1646444
Feyerabend, Paul K. (1978). Science in a free society. Lowe & Brydone Ltd.
Feyerabend, Paul K. “On the improvement of the sciences and the arts, and the possible identity of the two”. Proceedings of the Boston Colloquium for the Philosophy of Science 1964/1966, editado por R. S. Cohen y M. W. Wartofsky. Springer, 1967, pp. 387–415. https://doi.org/10.1007/978-94-010-3508-8_22
Floridi, Luciano, et al. “AI4People—An ethical framework for a good AI society: Opportunities, risks, principles, and recommendations”. Minds and Machines, vol. 28, núm. 4, 2018, pp. 689–707. https://doi.org/10.1007/s11023-018-9482-5
Goodall, Noah J. “Away from trolley problems and toward risk management”. Applied Artificial Intelligence, vol. 30, núm. 8, 2016, pp. 810–821. https://doi.org/10.1080/08839514.2016.1229922
Goodfellow, Ian J., et al. “Generative adversarial nets”. Advances in Neural Information Processing Systems, vol. 27, 2014. https://doi.org/10.48550/arXiv.1406.2661
Haas, Julia, et al. “A roadmap for evaluating moral competence in Large Language Models”. Nature, vol. 650, 2026, pp. 565–573. https://doi.org/10.1038/s41586-025-10021-1
Haugeland, John. Artificial intelligence: The Very Idea. MIT Press, 1989.
Himmelreich, Johannes. “Never mind the trolley: The ethics of autonomous vehicles in mundane situations”. Ethical Theory and Moral Practice, vol. 21, 2018, pp. 669–684. https://doi.org/10.1007/s10677-018-9896-4
Longino, Helen E. “Beyond ‘bad science’”. Science, Technology, & Human Values, vol. 8, núm. 1, 1983, pp. 7–17.
Matthews, Kyle y Marie Lamensch. Wired for war: how authoritarian states are weaponizing AI against the west. Konrad-Adenauer-Stiftung e.V., 2025.
McCarthy, John, et al. “A proposal for the Dartmouth summer research project on artificial intelligence”. 1955 http://jmc.stanford.edu/articles/dartmouth/dartmouth.pdf
Miller, Seumas. “Cognitive warfare: An ethical analysis”. Ethics and Information Technology, vol. 25, article number 46, 2023. https://doi.org/10.1007/s10676-023-09717-7
Moore, George Edward. Principia ethica. Cambridge University Press, 2002.
NATO STO. “Cognitive warfare: NATO Chief Scientist research report”. NATO, 2025. https://www.nato.int/content/dam/nato/webready/documents/sto/chief-scientist-report-cognitive-warfare.pdf
NATO. “NATO’s approach to counter information threats”. NATO, 2025. https://www.nato.int/en/what-we-do/wider-activities/natos-approach-to-counter-information-threats
Newen, Albert, et al. The Oxford Handbook of 4E Cognition. Oxford University Press, 2018.
Oh, Soyoung y Vera Demberg. “Robustness of large language models in moral judgements”. Royal Society Open Science, vol. 12, núm. 4, 2025, pp. 241229. https://doi.org/10.1098/rsos.241229
Origgi, Gloria. Is trust an epistemological notion? Episteme, vol. 1, núm. 1, 2004, pp. 61–72. https://doi.org/10.3366/epi.2004.1.1.61
Parasuraman, Raja y Victor Riley. “Humans and automation: Use, misuse, disuse, abuse”. Human Factors, vol. 39, núm 2, 1997, pp. 230–253. doi.org/10.1518/001872097778543886
Pratzner, Philip R. “The current targeting process”. Targeting: the challenges of modern warfare, editado por P. A. L. Ducheine, M. N. Schmitt y F. P. B. Osinga. T.M.C. Asser Press, 2016, pp. 77-97. https://doi.org/10.1007/978-94-6265-072-5_4
Putnam, Hilary. The collapse of the fact/value dichotomy and other essays. Harvard University Press, 2002.
Rodilosso, Ermelinda. “Epistemic vs. moral: A differentiated approach to human-centered AI ethics”. Topoi, 2025. https://doi.org/10.1007/s11245-025-10324-y
Russell, Stuart y Peter Norvig. Artificial intelligence: a modern approach (4th ed.). Pearson, 2021.
Sajadieh, Sha, et al. The AI index 2026 annual report. AI Index Steering Committee, Institute for Human-Centered AI, Stanford University, 2026.
Sans Pinillos, Alger y David Casacuberta. “Remarks on the possibility of ethical reasoning in an artificial intelligence system by means of abductive models”. Model-based reasoning in science and technology: MBR 2018, editado por Á. Nepomuceno-Fernández, L. Magnani, F. Salguero-Lamillar, C. Barés-Gómez y M. Fontaine. Springer, 2019, pp. 318–333. https://doi.org/10.1007/978-3-030-32722-4_19
Sans Pinillos, Alger y Lorenzo Magnani. “How do we think about the unknown? The self-awareness of ignorance as a tool for managing the anguish of not knowing”. Embodied, extended, ignorant minds, editado por S. Arfini y L. Magnani. Springer, 2022. https://doi.org/10.1007/978-3-031-01922-7_9
Sans Pinillos, Alger y Vincent Costa. “Más allá de los datos: la transformación digital del museo tradicional”. Daimon Revista Internacional De Filosofía, vol. 90, 2023, pp. 81–94. https://doi.org/10.6018/daimon.563231
Sans Pinillos, Alger. “Apuntes sobre los aspectos de valor prescriptivo del razonamiento abductivo”. El jardín de senderos que se bifurcan y confluyen: filosofía, lógica y matemáticas, editado por D. P. Fernandes y R. Lopez-Orellana. Universidad de Valparaíso, 2020, pp. 143–155.
Sans Pinillos, Alger. “CELL (Contextual Ethics Level Layer) as an architecture for contextual ethical governance in dual-use defense technologies”. Clicking the pause: the role of transatlantic cooperation in artificial intelligence supervision, editado por F. Andrés Pérez y R. García Alonso. NATO Science for Peace and Security Series E: Human and Societal Dynamics, 2026, pp. 145-165. https://doi.org/10.3233/NHSDP260044
Sans Pinillos, Alger. “Horror vacui: Characterizing the experience of paradigm crisis through Magnani’s EC-model of abduction”. Scientific cognition, semiotics, and computational agents: essays in honor of Lorenzo Magnani. Vol. 2, editado por S. Arfini. Springer, 2025, pp. 171-192. https://doi.org/10.1007/978-3-031-96688-0_9
Sans Pinillos, Alger. “Neglected pragmatism: discussing abduction to dissolute classical dichotomies”. Foundations of Science, vol. 27, 2022, pp. 1107–1125. https://doi.org/10.1007/s10699-021-09817-x
Sans Pinillos, Alger. “Unpacking bad expectations: a framework for anticipation and social justice in the eco-cognitive paradigm”. Model-Based reasoning, abductive cognition, creativity. MBR 2023, editado por E. Ippoliti, L. Magnani y S. Arfini. Springer, 2024, pp. 254-270. https://doi.org/10.1007/978-3-031-69300-7_15
Schmitt, Michael N. y Eric Widmar. “The law of targeting”. Targeting: the challenges of modern warfare, editado por P. A. L. Ducheine, M. N. Schmitt y F. P. B. Osinga. T.M.C. Asser Press, 2016, pp. 121–145. https://doi.org/10.1007/978-94-6265-072-5_6
Sheridan, Thomas B. Telerobotics, automation, and human supervisory control. MIT Press, 1992.
Simon, Herbert A. Models of man: social and rational. Wiley, 1957.
Skitka, Linda, et al. “Does automation bias decision-making?” International Journal of Human-Computer Studies, vol. 51, núm 5, 1999, pp. 991–1006. https://doi.org/10.1006/ijhc.1999.0252
Vallverdú, Jordi y Alger Sans Pinillos. “The foundations of creativity: Human inquiry explained through the neuro-multimodality of abduction”. Handbook of abductive cognition, editado por Lorenzo Magnani. Springer, 2022, pp. 1-27. https://doi.org/10.1007/978-3-030-68436-5_71-1
Varela, Francisco J., et al. The embodied mind: cognitive science and human experience. MIT Press, 2017.
Vaswani, Ashish, et al. “Attention is all you need”. Advances in Neural Information Processing Systems, vol. 30, 2017. https://doi.org/10.48550/arXiv.1706.03762
Additional Files
Published
Issue
Section
License
Copyright (c) 2026 DISCOURSE STUDIES
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Since 2023, the journal has been published in open access under the CC BY-NC 4.0 Attribution/NonCommercial 4.0 International license, which allows you to share: copy and redistribute the material in any medium or format; and adapt: remix, transform, and build upon the material. Attribution: you must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in a reasonable manner, but not in any way that suggests that the licensor endorses you or your use. NonCommercial: you may not use the material for commercial purposes.
