Publicado
De fábulas y autómatas: reflexiones iniciales de la narrativa para la alineación de valores de la inteligencia artificial
Of Fables and Automata: Initial Reflections on the Narrative for Value Alignment of Artificial Intelligence
De fábulas e autômatos: reflexões iniciais sobre a narrativa para o alinhamento de valores da Inteligência Artificial
DOI:
https://doi.org/10.15446/rcpeha.n18.106078Palabras clave:
inteligencia artificial, inteligencia artificial general, inteligencia narrativa, narrativa, alineación de valores, ética, AlphaGo, ChatGPT (es)artificial intelligence, artificial general intelligence, narrative intelligence, narrative, value alignment, ethics, AlphaGo, ChatGPT (en)
inteligência artificial, inteligência artificial geral, inteligência narrativa, narrativa, alinhamento de valores, ética, AlphaGo, ChatGPT (pt)
Descargas
el artículo aborda la necesidad de alinear la inteligencia artificial (IA) con los valores humanos ante el desarrollo de la IA general, capaz de igualar o superar la inteligencia humana. Se plantea que la narrativa puede ser fundamental para infundir principios éticos en las IA, dado su rol en la interpretación del mundo y la representación de experiencias humanas. Mediante una revisión bibliográfica de fuentes académicas, el estudio explora la relación entre la IA y la narrativa, enfocándose en la identidad narrativa, el conocimiento tácito y la experiencialidad narrativa. Se analizan casos como AlphaGo y ChatGPT para ilustrar la diferenciación entre IA estrecha y general, y se examina cómo las filosofías utilitarista y deontológica pueden influir en la ética de la IA. Los hallazgos identifican tres perspectivas fundamentales: la identidad narrativa, que configura la identidad a través de historias; el conocimiento tácito, que resalta el aprendizaje implícito y la experiencia intuitiva; y la experiencialidad narrativa, que subraya la importancia de la corporeidad y la vivencia temporal en la comprensión de relatos. La narrativa emerge como un enfoque viable para la alineación ética de la IA, aunque enfrenta desafíos relacionados con la experiencia vital y la corporeidad, elementos cruciales para una auténtica comprensión narrativa.
the article addresses the need to align artificial intelligence (AI) with human values in the face of the development of general AI capable of matching or surpassing human intelligence. It argues that narrative may be central to instilling ethical principles in AI, given its role in interpreting the world and representing human experiences. Through a literature review of academic sources, the study explores the relationship between AI and narrative, focusing on narrative identity, tacit knowledge, and narrative experientiality. Cases such as AlphaGo and ChatGPT are analyzed to illustrate the differentiation between narrow and general AI, and it examines how utilitarian and deontological philosophies may influence AI ethics. The findings identify three fundamental perspectives: narrative identity, which shapes identity through stories; tacit knowledge, which highlights implicit learning and intuitive experience; and narrative experientiality, which stresses the importance of embodiment and temporal liveness in understanding narratives. Narrative emerges as a viable approach to the ethical alignment of AI, although it faces challenges related to life experience and embodiment, crucial elements for authentic narrative understanding.
o artigo aborda a necessidade de alinhar a inteligência artificial (IA) aos valores humanos diante do desenvolvimento de uma IA geral capaz de igualar ou superar a inteligência humana. Argumenta-se que a narrativa pode ser fundamental para incutir princípios éticos na IA, dada sua função de interpretar o mundo e representar as experiências humanas. Por meio de uma revisão da literatura de fontes acadêmicas, o estudo explora a relação entre IA e narrativa, concentrando-se na identidade narrativa, no conhecimento tácito e na experiência narrativa. Casos como o AlphaGo e o ChatGPT são analisados para ilustrar a diferenciação entre IA restrita e geral, e o estudo examina como as filosofias utilitarista e deontológica podem influenciar a ética da IA. As descobertas identificam três perspectivas principais: identidade narrativa, que molda a identidade por meio de histórias; conhecimento tácito, que destaca o aprendizado implícito e a experiência intuitiva; e experiencialidade narrativa, que enfatiza a importância da incorporação e da experiência temporal na compreensão das narrativas. A narrativa surge como uma abordagem viável para o alinhamento ético da IA, embora enfrente desafios relacionados à experiência de vida e à incorporação, elementos cruciais para a compreensão autêntica da narrativa.
Referencias
Bibliografía
Fuentes secundarias
“Introducing ChatGPT”. OpenAI (blog), 30 de noviembre de 2022. https://openai.com/blog/chatgpt
Basilan, Marvie. “Elon Musk Predicts ‘Digital Superintelligence’ To Arrive In 5-6 Years, Launches AI Company”. International Business Times, 13 de julio de 2023. https://www.ibtimes.com/elon-musk-predicts-digital-superintelligencearrive-5-6-years-launches-ai-company-3704835
Bostrom, Nick. Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press, 2014.
Busacchi, Vinicio. “On Narrative Identity”. En 5th International Multidisciplinary Scientific Conference on Social Sciences and Arts. Conference Proceedings, vol. 5. Ancience Science, vol 2.2, 555-563. Sofía: SGEM, 2018.
Campbell, Joseph. The Hero with a Thousand Faces. Princeton: Princeton University Press, 2004.
Caracciolo, Marco. “Experientiality”. En The living handbook of narratology, editado por Peter Hühn, Jan-Christoph Meister, John Pier y Wolf Schmid. Hamburgo: Hamburg University Press, 2014. https://www-archiv.fdm.uni-hamburg.de/lhn/node/102.html
Coeckelbergh, Mark. Imagination and Principles. An Essay on the Role of Imagination in Moral Reasoning. Londres: Palgrave Macmillan, 2007. https://doi.org/10.1057/9780230589803
Dreyfus, Hubert L., Stuart E. Drey-fus y Lotfi A. Zadeh, Mind over Machine: The Power of Human Intuition and Expertise in the Era of the Computer. Nueva York: The Free Press, 1986.
Fernyhough, Charles. Pieces of Light: The New Science of Memory. Londres: Profile Books, 2012.
Fjelland, Ragnar. “Why general artificial intelligence will not be realized”. Humanities and Social Sciences Communications 7, no. 10 (2020). https://doi.org/10.1057/s41599-020-0494-4
Fludernik, Monika. Towards a natural Narratology. Londres: Taylor & Francise-Library, 1996.
Fludernik, Monika. “Natural Narratology and Cognitive Parameters”. En Narrative Theory and the Cognitive Sciences, editado por David Herman, 243-267. Stanford: Center for the Study of Language and Information Publications, 2003.
Gabriel, Iason. “Artificial Intelligence, Values, and Alignment”. Minds and Machines 30 (2020): 411-437. https://doi.org/10.1007/s11023-020-09539-2
Harari, Yuval Noah. Sapiens. De animales a dioses: Una breve historia de la humanidad. Barcelona: Debate, 2014.
Herman, David. “Scripts, Sequences, and Stories: Elements of a Postclassical Narratology”. Publications of the Modern Language Association of America 112, no. 5 (1997): 1046-1059. https://doi.org/10.2307/463482
Hibbard, Bill. Super-Intelligent Machines. Nueva York: Springer, 2002. https://doi.org/10.1007/978-1-4615-0759-8
Jayanti, Vikram. dir., Game Over: Kasparov and the Machine, 2003.
Kiernan, Patrick. “Which is greater? The number of atoms in the universe or the number of chess moves?”. Nationals Museums Liverpool (página web), s.f. https://www.liverpoolmuseums.org.uk/stories/which-greater-number-of-atomsuniverse-or-number-of-chess-moves
Kohs, Greg, dir. AlphaGo. 2017. YouTube, 13 de diciembre de 2017. https://www.youtube.com/watch?v=WXuK6gekU1Y
Korteling, Johan Egbert (Hans), Gillian Van De Boer-Visschedijk, Romy A.M. Blankendaal, Rudy Boonekamp y Aletta Eikelboom. “Human- versus Artificial Intelligence”. Frontiers in Artificial Intelligence 4 (2021): 1-13. https://doi.org/10.3389/frai.2021.622364
Mateas, Michael y Phoebe Sengers, “Narrative Intelligence”. En Narrative Intelligence, editado por Michael Mateas y Phoebe Sengers, 1-25. Ámsterdam y Philadelphia: John Benjamins Publishing Company, 2003.
McAdams, Dan P. “‘First we invented stories, then they changed us’: The Evolution of Narrative Identity”. Evolutionary Studies in Imaginative Culture 3, no. 1 (2019): 1-18. https://doi.org/10.26613/esic.3.1.110
Merleau-Ponty, Maurice. Fenomenología de la percepción. Barcelona: Planeta DeAgostini, [1945], 1993.
Polanyi, Michael. Personal Knowledge: Towards a Post-Critical Philosophy. Londres: Routledge, [1958] 1998. Reijers, Wessel y Mark Coeckelbergh. Narrative and technology ethics. Cham: Palgrave Macmillian, 2020.
Ricoeur, Paul. “La vida: un relato en busca de narrador”. Ágora: Papeles de filosofía 25, no. 2 (2006): 9-22.
Riedl, Mark O. y Brent Harrison. “Using Stories to Teach Human Values to Artificial Agents”. AAAI Workshop: AI, Ethics, and Society (2016). https://www.semanticscholar.org/paper/Using-Stories-to-Teach-Human-Values-to-Artificial-Riedl-Harrison/33b53abdf2824b2cb0ee083c284000df4343a33e
Rogers, Benjamin A., Herrison Chicas, John Michael Kelly, Emily Kubin, Michael S. Christian, Frank J. Kachanoff, Jonah Berger, Curtis Puryear, Dan P. McAdams y Kurt Gray. “Seeing your life story as a Hero’s Journey increases meaning in life”. Journal of Personality and Social Psychology 125, no. 4 (2023): 752-778. https://doi.org/10.1037/pspa0000341
Ryan, Marie-Laure. “Narrative mapping as cognitive activity and as active participation in storyworlds”. Frontiers of Narrative Studies 4, no. 2 (2018): 232-247. https://doi.org/10.1515/fns-2018-0020
Shevlin, Henry, Karina Vold, Matthew Crosby y Marta Halina. “The limits of machine intelligence: Despite progress in machine intelligence, artificial general intelligence is still a major challenge”. EMBO Reports 20 (2019):e49177. https://doi.org/10.15252/embr.201949177
Soares, Nate. “The Value Learning Problem”, Artificial Intelligence Safety and Security (2018). https://www.semanticscholar.org/paper/The-Value-Learning-Problem-Soares/f042e8db015e31b13a8aa2a8b814ff745b3f4f49
Yalçın, Özge Nilay y Steve Dipaola. “Modeling empathy: building a link between affective and cognitive processes”. Artificial Intelligence Review 53, no. 1 (2020): 2983-3006. https://doi.org/10.1007/s10462-019-09753-0
Zhang, Chaoning, Chenshuang Zhang, Chenghao Li, Yu Qiao, Sheng Zheng, Sumit Kumar Dam, Mengchun Zhang, Jung Uk Kim, Seong Tae Kim, Jinwoo Choi, Gyeong-Moon Park, Sung-Ho Bae, Lik-Hang Lee, Pan Hui, In So Kweon y Choong Seon Hong. “One Small Step for Generative AI, One Giant Leap for AGI: A Complete Survey on ChatGPT in AIGC Era”. arXiv:2304.06488v1 (2023). https://doi.org/10.48550/arXiv.2304.06488