Српска политичка мисао 6/2025

RISK AND RESPONSIBILITY AT THE FRONTIER OF AI: A THEMATIC ANALYSIS OF DEEP LEARNING PIONEERS’ PERSPECTIVES ON ARTIFICIAL INTELLIGENCE THREATS AND GOVERNANCE

Аутори
УДКUDC 327:316.776:004.774(100) Стране137-165 Година2025
Преузми PDF 10.5937/spm94-59969
Преглед истраживања

Сажетак

As artificial intelligence (AI) reshapes global societies, understanding its associated risks and governance imperatives is of urgent social importance. This study fills a critical gap by systematically analyzing extended interviews with Geoffrey Hinton, Yoshua Bengio, and Yann LeCun – to elucidate their firsthand perspectives on AI’s existential, ethical, social, and governance challenges. Employing qualitative thematic analysis across six longitudinal interview transcripts, the research identifies both convergences and divergences: Hinton and Bengio strongly emphasize existential threats, superintelligence hazards, AI weapons risks, and the need for robust global regulation, while LeCun expresses technological optimism and favors decentralized, open development. All acknowledge economic disruption, misuse of potential, and fractures in democratic discourse. The study’s findings reveal that expert opinion on AI risk is far from monolithic and highlight actionable, innovative governance proposals, from regulated compute access to “diversity engines” in social media feeds. Implications include the necessity for adaptive, internationally coordinated AI governance and greater professional accountability among developers. Limitations include a focus on elite, Anglophone experts and inherent subjectivity in qualitative coding. Future research should expand to multi-stakeholder and cross-national perspectives, and test proposed regulatory frameworks in real-world contexts, addressing the ongoing evolution of risk as AI permeates new domains.

Извори и литература

Референце

  1. Alipour, Shayan, Alessandro Galeazzi, Emanuele Sangiorgio, Michele Avalle, Ljubisa Bojic, Matteo Cinelli, and Walter Quattrociocchi. 2024. “Cross-Platform Social Dynamics: An Analysis of ChatGPT and COVID-19 Vaccine Conversations.” Scientific Reports 14 (1): 2789. DOI:10.1038/s41598-024-53124-x
  2. Altmann, Jürgen, and Frank Sauer. 2017. “Autonomous Weapon Systems and Strategic Stability.” Survival 59: 117–42. DOI:10.1080/00396338.2017.1375263
  3. Autor, David H. 2025. “Why Are There Still So Many Jobs? The History and Future of Workplace Automation.” Journal of Economic Perspectives 29 (3): 3–30
  4. Bengio_BBC. 2025. “The Worst Case Scenario Is Human Extinction – Godfather of AI on Rogue AI.” BBC YouTube. https://youtu.be/c4Zx849dOiY
  5. Bengio_WSF. 2024. “Why a Forefather of AI Fears the Future.” World Science Festival, YouTube. https://youtu.be/KcbTbTxPMLc
  6. Bessen, James E. 2018. “AI and Jobs: The Role of Demand.” NBER Working Paper. https://www.nber.org/papers/w24235
  7. Bodroža, Bojana, Bojana M. Dinić, and Ljubiša Bojić. 2024. “Personality Testing of Large Language Models: Limited Temporal Stability, but Highlighted Prosociality.” Royal Society Open Science 11: 240180. DOI:10.1098/rsos.240180
  8. Bojic, Ljubisa. 2022. “Metaverse through the Prism of Power and Addiction.” European Journal of Futures Research 10 (1): 22. DOI:10.1186/s40309-022-00208-4
  9. Bojic, Ljubisa. 2024. “AI Alignment: Assessing the Global Impact of Recommender Systems.” Futures 160: 103383. DOI:10.1016/j.futures.2024.103383
  10. Bojic, Ljubisa, and Jean-Louis Marie. 2017. “Addiction to Old versus New Media.” Srpska politička misao 56 (2): 33–48. DOI:10.22182/spm.5622017.2
  11. Bojić, Ljubiša, Irena Stojković, and Zorana Jolić Marjanović. 2024. “Signs of Consciousness in AI.” Humanities and Social Sciences Communications 11: 1631. DOI:10.1057/s41599-024-04154-3
  12. Bojić, Ljubiša, Matteo Cinelli, Dubravko Ćulibrk, and Boris Delibašić. 2024. “CERN for AI.” European Journal of Futures Research 12: 15. DOI:10.1186/s40309-024-00238-0
  13. Bojic, Ljubisa, Milos Agatonović, and Jelena Guga. 2024. “The Immersion in the Metaverse.” In Augmented and Virtual Reality in the Metaverse. Springer. DOI:10.1007/978-3-031-57746-8_11
  14. Bojić, Ljubiša et al. 2025. “Comparing Large Language Models and Human Annotators.” Scientific Reports 15: 11477. DOI:10.1038/s41598-025-96508-3
  15. Bojić, Ljubiša, Predrag Kovačević, and Milan Čabarkapa. 2025. “Does GPT-4 Surpass Human Performance in Linguistic Pragmatics?” Humanities and Social Sciences Communications 12: 794. DOI:10.1057/s41599-025-04912-x
  16. Bostrom, Nick. 2014. Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press
  17. Braun, Virginia, and Victoria Clarke. 2006. “Using Thematic Analysis in Psychology.” Qualitative Research in Psychology 3 (2): 77–101
  18. Brundage, Miles et al. 2018. “The Malicious Use of Artificial Intelligence.” arXiv:1802.07228. DOI:10.48550/arXiv.1802.07228
  19. Brynjolfsson, Erik, and Andrew McAfee. 2014. The Second Machine Age. New York: W. W. Norton
  20. Buolamwini, Joy, and Timnit Gebru. 2018. “Gender Shades.” Proceedings of Machine Learning Research 81: 1–15
  21. Cath, Corinne. 2018. “Governing Artificial Intelligence.” Philosophical Transactions of the Royal Society A 376: 20180080
  22. Chui, Michael, James Manyika, and Mehdi Miremadi. 2016. “Where Machines Could Replace Humans.” McKinsey Quarterly
  23. De Filippi, Primavera, Samer Hassan, and Roberto Zicari. 2023. “Smart Contracts Meet Legal Responsibility.” Stanford Journal of Blockchain Law & Policy 6 (1): 77–104
  24. Doshi-Velez, Finale, and Been Kim. 2017. “Towards a Rigorous Science of Interpretable Machine Learning.” arXiv:1702.08608
  25. Esteva, Andre et al. 2017. “Dermatologist-Level Classification of Skin Cancer.” Nature 542: 115–118
  26. Feldstein, Steven. 2019. “The Global Expansion of AI Surveillance.” Carnegie Endowment
  27. Floridi, Luciano et al. 2018. “AI4People.” Minds and Machines 28 (4): 689–707
  28. Gabriel, Iason. 2020. “Artificial Intelligence, Values, and Alignment.” Minds and Machines 30 (3): 411–437
  29. Helberger, Natali. 2019. “On the Democratic Role of News Recommenders.” Digital Journalism 7 (8): 993–1012
  30. Jobin, Anna, Marcello Ienca, and Effy Vayena. 2019. “The Global Landscape of AI Ethics Guidelines.” Nature Machine Intelligence 1 (9): 389–399
  31. Korinek, Anton, and Joseph E. Stiglitz. 2018. “Artificial Intelligence and Income Distribution.” In Economics of Artificial Intelligence. Chicago: University of Chicago Press
  32. Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. 2017. “ImageNet Classification.” Communications of the ACM 60 (6): 84–90
  33. Noble, Safiya Umoja. 2018. Algorithms of Oppression. New York: NYU Press
  34. O’Neil, Cathy. 2016. Weapons of Math Destruction. New York: Broadway Books
  35. Russell, Stuart. 2019. Human Compatible. New York: Viking
  36. Zuboff, Shoshana. 2019. The Age of Surveillance Capitalism. New York: PublicAffairs.
Из истог броја

Повезане публикације