Keywords
AI apologies
romantic authenticity
gender differences
campus relationships
Abstract
This study investigates "digital surrogacy", the use of AI tools like ChatGPT as substitutes for human emotional expression in romantic apologies among University of Ibadan students. Anchored in Peng's (2020) Authenticity Model of Computer-Mediated Communication, the research examines whether machine-generated apologies undermine relational authenticity in cross-gender campus relationships between Mellanby Hall males and Queen's Hall females. Employing a quantitative cross-sectional survey (N=384), structured Likert-scale questionnaires assessed authenticity perceptions across romantic scenarios via Google Forms distributed through hall WhatsApp groups.
Findings reveal human apologies significantly outperformed AI-generated ones (M=4.33 vs 2.37; t=22.67, p<0.001), confirming AI's emotional deficits. Queen's females demonstrated stronger rejection of digital surrogacy (M=3.98 vs males M=3.67), prioritising relational
warmth, while males favoured pragmatic resolution. Thematic analysis identified "lack of human heart" (77.6%) as the primary authenticity barrier. Recommendations include developing culturally-sensitive hybrid AI tools incorporating pidgin emotionality, relationship counselling emphasising human authenticity, and university policies guiding ethical AI use in intimate communications to preserve campus love dynamics. The study concludes AI-mediated apologies erode romantic trust, particularly among female respondents who value emotional congruence.
References
1. Dahlin, E., et al. (2021). Mind the gap! On the future of AI research. Humanities and Social Sciences Communications, 8(1), Article 88. https://doi.org/10.1057/s41599-021-00750-9
2. Doyle, A., et al. (2023). Apology psychology: Breaking gender stereotypes leads to more effective communication. University of Arizona Eller College of Management. https://eller.arizona.edu/news/apology-psychology-breaking-gender-stereotypes
3. Glikson, E., & Asscher, O. (2023). AI-mediated apology in a multilingual work context: Implications for perceived authenticity and willingness to forgive. Computers in Human Behaviour, 140, Article 107592. https://doi.org/10.1016/j.chb.2022.107592
4. Kirk, C. P., Galla, Z., & Solomon, M. R. (2025). The AI-authorship effect: Understanding authenticity, moral responsibility, and consumer responses. Journal of Business Research, 170, Article 114300. https://doi.org/10.1016/j.jbusres.2024.114300
5. Lawong, K. Y. (2023). Gender and communicative styles in apologising among Nigerian students. SAR Journal of Applied Linguistics and Literature. https://sarpublication.com
6. Lim, J. Y., Zhang, Y., & Wang, L. (2025). How warm-versus competent-toned AI apologies affect human perceptions of sincerity and forgiveness. Computers in Human Behaviour, Article 108208. https://doi.org/10.1016/j.chb.2025.108208
7. Microsoft. (2025). AI Diffusion Report 2025. [Various sources]
8. NCC Nigeria. (2024). Baseline study on the level of digital skills in Nigeria. https://ncc.gov.ng
9. Peng, T. Q. (2020). Authenticity model of (mass-oriented) computer-mediated communication: Integrating theory and method toward greater scientific rigour. Journal of Computer-Mediated Communication, 25(1), 60–79. https://doi.org/10.1093/jcmc/zmz029
10. Stanford HAI. (2025). The 2025 AI Index Report. https://hai.stanford.edu/ai-index/2025-ai-index-report
11. Wang, X., & Li, Y. (2024). A critical review of apology in AI systems. arXiv preprint arXiv:2412.15787. https://arxiv.org/abs/2412.15787
12. World Bank. (2025). Who on Earth is using generative AI? Global patterns. https://openknowledge.worldbank.org
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2026 Author