Ethical Challenges in AI Emotional Interaction: Mechanisms of Emotional Dependence and Governance Pathways
DOI:
https://doi.org/10.66581/6d1kt879Keywords:
AI emotional interaction, emotional dependence, multimodal affective computing, mental health, explainabilityAbstract
This paper focuses on the ethical risks of AI emotional interaction, particularly the formation mechanisms and governance pathways of “emotional dependence.” By reviewing foundational technologies such as sentiment analysis, speech recognition, and NLP, and using LSTM based long term memory reinforcement, Bayesian emotion state transitions, data feedback loops, and real time physiological signal monitoring as the main analytic thread, it reveals how platforms, through multimodal personalization and just in time adaptation, construct “parasocial relationships” that amplify users’ behavioral addiction, anthropomorphic cognitive biases, and social alienation. Drawing on cases including the use of Replika during the pandemic, instances of digital immortality, and AI personas on social platforms, the paper identifies the mental health risks, the weakening of user autonomy, and ambiguity in responsibility attribution induced by emotional dependence. In response, it proposes a system level governance pathway: addiction mitigation design centered on session duration controls and emotional intensity thresholds; enhanced explainability via attention visualization; tiered, scenario aware, and dynamically adjustable permission management calibrated by age and psychological resilience; and mandatory ethics reviews with third party assessments across the full lifecycle. The paper builds a “mechanism–risk–governance” analytical framework and advocates a shift from anthropomorphic to assistive design paradigms, strengthening transparency and accountability to promote prudent innovation and the healthy development of affective AI.
References
Agrawal, R., & Pandey, N. (2024). Developing Rapport Between Humans and Machines: Emotionally Intelligent AI Assistants. International Journal for Research in Applied Science and Engineering Technology, 12(3), 1473-1480. doi:10.22214/ijraset.2024.59015
Amara, K., Kerdjidj, O., & Ramzan, N. (2023). Emotion Recognition for Affective Human Digital Twin by Means of Virtual Reality Enabling Technologies. Ieee Access, 11, 74216-74227. doi:10.1109/access.2023.3285398
Anupama, A., Save, V., Biswas, A., Kumar, V., Lourens, M., Madhura, K., & Sharma, A. (2023). Deep Learning Approach For Emotions Detection. Paper presented at the E3S Web of Conferences.
Bekkemoen, Y. (2024). Explainable reinforcement learning (XRL): a systematic literature review and taxonomy. Machine Learning, 113(1), 355-441.
Chandra, D. P., & Rajarajeswari, S. (2019). An optimized hybrid neural network model for detecting depression among twitter users. International Journal of Innovative Technology and Exploring Engineering, 8(10), 2781-2795.
Chen, J., Lin, M., Zheng, Z., & Lv, H. (2023). Method and Data Analysis Based on ECG Signal for Emotion Recognition. 63. doi:10.1117/12.3009419
Cheng, J. (2023). Innovation of Fresh Agricultural Products E-Commerce Operation Mode Based on Artificial Intelligence and Machine Learning Model. Applied Mathematics and Nonlinear Sciences, 9(1).
Corrêa, R. S., Marques, A., Neufeld, C. B., de Almeida, A. M., & de Matos, M. G. (2022). Desenvoltura social, regulação parental e atividade física em adolescentes: um estudo de corte. Research, Society and Development, 11(8), e30111830818-e30111830818.
Dai, L., & Chen, P. (2019). Reasons for the popularity of Tik Tok, the shortages and the ways forward. Paper presented at the 2019 3rd International Seminar on Education, Management and Social Sciences (ISEMSS 2019).
Guo, R., Guo, H., Wang, L., Chen, M., Dong, Y., & Li, B. (2024). Development and Application of Emotion Recognition Technology — A Systematic Literature Review. BMC Psychology, 12(1). doi:10.1186/s40359-024-01581-4
Huang, S., Lai, X., Ke, L., Li, Y., Wang, H., Zhao, X., . . . Wang, Y. (2024). AI Technology panic—is AI dependence bad for mental health? A cross-lagged panel model and the mediating roles of motivations for AI use among adolescents. Psychology Research and Behavior Management, 1087-1102.
Kalam, K. T., Rahman, J. M., Islam, M. R., & Dewan, S. M. R. (2024). ChatGPT and mental health: Friends or foes? Health Science Reports, 7(2), e1912.
Khawaja, Z., & Bélisle-Pipon, J.-C. (2023). Your robot therapist is not your therapist: understanding the role of AI-powered mental health chatbots. Frontiers in Digital Health, 5, 1278186.
Kołakowska, A., Szwoch, W., & Szwoch, M. (2020). A review of emotion recognition methods based on data acquired via smartphone sensors. Sensors, 20(21), 6367.
Lei, X. (2022). The impact of emotion management ability on learning engagement of college students during COVID-19. Frontiers in psychology, 13, 967666.
Lewis, P. A., Critchley, H. D., Smith, A., & Dolan, R. J. (2005). Brain mechanisms for mood congruent memory facilitation. Neuroimage, 25(4), 1214-1223.
Li, M., Zhu, Y., Shen, Y., & Angelova, M. (2023). Clustering-enhanced stock price prediction using deep learning. World Wide Web, 26(1), 207-232.
López, A. M., León, E., & Quintero Montoya, O. L. (2022). Biometric monitoring of emotions and behaviors. In (Vol. 16, pp. 1079761): Frontiers Media SA.
Quispe, K. G. M., Utyiama, D. M. S., Santos, E. M. d., Oliveira, H. A., & Souto, E. (2022). Applying Self-Supervised Representation Learning for Emotion Recognition Using Physiological Signals. Sensors, 22(23), 9102. doi:10.3390/s22239102
Rodríguez-García, M. I., Carrasco-García, M. G., González-Enrique, J., Ruiz-Aguilar, J. J., & Turias, I. J. (2023). Long short-term memory approach for short-term air quality forecasting in the Bay of Algeciras (Spain). Sustainability, 15(6), 5089.
Rodríguez Reséndiz, H., & Rodríguez Reséndiz, J. (2024). Digital resurrection: Challenging the boundary between life and death with artificial intelligence. Philosophies, 9(3), 71.
Sethi, S. S., & Jain, K. (2024). AI Technologies for Social Emotional Learning: Recent Research and Future Directions. Journal of Research in Innovative Teaching & Learning, 17(2), 213-225. doi:10.1108/jrit-03-2024-0073
Shevlin, H. (2024). All too human? Identifying and mitigating ethical risks of Social AI. Law, Ethics & Technology, 1(2), 1-22.
Shuai, S. (2023). xploring the direction of the English translation of environmental protection articles based on the robot cognitive-emotional interaction model. 3 c TIC: cuadernos de desarrollo aplicados a las TIC, 12(1), 222-246.
Steinberg, H., Mollborn, S., & Pace, J. (2024). “Mature Enough to Handle it?”: Gendered Parental Interventions in and Adolescents’ Reactions to Technology Use During the Pandemic. Journal of Family Issues, 45(1), 237-258.
Sun, L., Fu, S., & Wang, F. (2019). Decision tree SVM model with Fisher feature selection for speech emotion recognition. EURASIP Journal on Audio, Speech, and Music Processing, 2019(1), 1-14.
Wang, P., Gan, X., Li, H., & Jin, X. (2023). Parental marital conflict and internet gaming disorder among Chinese adolescents: The multiple mediating roles of deviant peer affiliation and teacher-student relationship. PloS one, 18(1), e0280302.
Xu, W., Ge, L., & Gao, Z. (2021). Human-AI interaction: An emerging interdisciplinary domain for enabling human-centered AI. arXiv preprint arXiv:2112.01920.
Xygkou, A., Siriaraya, P., Covaci, A., Prigerson, H. G., Neimeyer, R., Ang, C. S., & She, W.-J. (2023). The" Conversation" about Loss: Understanding How Chatbot Technology was Used in Supporting People in Grief. Paper presented at the Proceedings of the 2023 CHI conference on human factors in computing systems.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Zhiyong Song (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
1.Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY 4.0). This allows others to share the work (copy and redistribute) and adapt the work (remix, transform, and build upon) for any purpose, even commercially, provided the original work is properly cited.
2.Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
3.Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
