A Comparative Study of Attitudinal Resources in Human and AI-Generated TED-Style Leadership Talks
DOI:
https://doi.org/10.57125/FS.2026.09.20.01Keywords:
appraisal theory, attitude system, interpersonal function, TED talks, AI-generated text, human-machine comparison.Abstract
This study compares the use of attitudinal resources in human TED talks and AI-generated TED-style texts from the perspective of appraisal theory. A corpus of 20 human TED talks on leadership and 20 AI-generated texts was constructed and manually annotated with UAM CorpusTool. The analysis focused on the distribution and polarity of affect, judgment, and appreciation resources. The findings show some observable differences between the two types of texts. Human texts draw more on affective resources, suggesting that human speakers in this corpus tend to construct evaluation through personal feelings, lived experience, and emotional involvement. In contrast, AI-generated texts show a more balanced distribution of the three resource types, with judgment taking a slightly higher proportion. In terms of polarity, the AI corpus contained a higher proportion of positive resources. In contrast, the human corpus contained a higher proportion of negative resources related to failure, doubt, and difficulty. This contrast was especially evident in judgment resources, in which negative judgment resources occurred less frequently in the AI corpus. Overall, the quantitative results and qualitative examples suggest that human texts in this corpus are more often grounded in personal experience and social relations. In contrast, AI-generated texts are more often characterised by generalised, instructive evaluative formulations.
References
Almurashi, W. A. (2016). An introduction to Halliday's systemic functional linguistics. Journal for the Study of English Linguistics, 4(1), 70–80.
Cheng, S. (2024). A review of interpersonal metafunction studies in systemic functional linguistics (2012–2022). Journal of World Languages, 10(3), 623–667.
Dong, J., & Zhang, M. (2026). Stance beyond words: How TED speakers construct stance through multimodal semiotic resources. English for Specific Purposes, 81, 150–168. https://doi.org/10.1016/j.esp.2025.10.001
Duan, Y., & Degand, L. (2024). Attitudinal resources in academic talks: A corpus-based analysis across languages. Corpus Pragmatics, 8(4), 335–358. https://doi.org/10.1007/s41701-024-00171-4
Flaih, M. A. A. K. N. (2026). Human vs. machine: A comparative study of traditional and AI-generated literary texts. Tasnim International Journal for Human, Social and Legal Sciences, 5(1), 152–163. https://doi.org/10.56924/tasnim.s1.2026/10
Gao, X. (2024). Positive discourse analysis of TED artificial intelligence education speech under attitude system. Journal of Higher Education Research, 5(5), 442–444. https://doi.org/10.32629/jher.v5i5.3057
Georgiou, G. P. (2025). Differentiating between human-written and AI-generated texts using automatically extracted linguistic features. Information, 16(11), Article 979. https://doi.org/10.3390/info16110979
Gillings, M., Kohn, T., & Mautner, G. (2025). The rise of large language models: Challenges for critical discourse studies. Critical Discourse Studies, 22(6), 625–641. https://doi.org/10.1080/17405904.2024.2373733
Gong, X. (2024). An interpersonal function analysis of English climate speeches. Modern English, (18), 96–98.
Halliday, M. A. K. (2004). An introduction to functional grammar (3rd ed.). Hodder Education.
Halliday, M. A. K., & Matthiessen, C. M. I. M. (2013). Halliday's introduction to functional grammar (4th ed.). Routledge.
Huo, H., & Zhang, Y. (2023). An analysis of the interpersonal function of Joe Biden's inaugural speech. International Journal of Languages, Literature and Linguistics, 9(3), 204–207.
Križan, A., & Barbič, A. (2025). Appraisal analysis and AI chatbots: Do we even need humans? ELOPE: English Language Overseas Perspectives and Enquiries, 22(1), 35–52. https://doi.org/10.4312/elope.22.1.35-52
Kumar, S., Tiwari, S., Prasad, R., Rana, A., & Arti, M. K. (2024). Comparative analysis of human and AI generated text. In 2024 11th International Conference on Signal Processing and Integrated Networks (SPIN) (pp. 168–173). IEEE. https://doi.org/10.1109/SPIN60856.2024.10511301
Mao, Y. (2025). An analysis of a TED speech from the perspective of interpersonal meta-function. International Journal of Linguistics, Literature and Translation, 8(3), 16–21. https://doi.org/10.32996/ijllt.2025.8.3.4
Martin, J. R., & White, P. R. R. (2005). The language of evaluation: Appraisal in English. Palgrave Macmillan.
Mindner, L., Schlippe, T., & Schaaff, K. (2023). Classification of human- and AI-generated texts: Investigating features for ChatGPT. In International Conference on Artificial Intelligence in Education Technology (pp. 152–170). Springer Nature Singapore.
Ming, Z. (2025). Attitudinal orientation in the translation of the 2025 New Year address from the perspective of appraisal theory: A corpus-based sentiment analysis of the attitude system. Language and Culture Studies, 33(3), 199–202.
Phan, L. T. (2025). Tracing the development of research on appraisal theory within systemic functional linguistics (2003–2025): A bibliometric analysis. Language Teaching Research Quarterly, 52, 87–109. https://doi.org/10.32038/ltrq.2025.52.05
Tang, Z., & Shen, W. (2025). Attitude resources of “a community with a shared future for mankind” in Singaporean news reports: A two-level sentiment analysis based on appraisal theory. Journalism & Communication Research, 32(12), 103–120, 123.
Tengler, K., & Brandhofer, G. (2025). Exploring the difference and quality of AI-generated versus human-written texts. Discover Education, 4(1), 113. https://doi.org/10.1007/s44217-025-00529-z
Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., & Peterson, S. J. (2008). Authentic leadership: Development and validation of a theory-based measure. Journal of Management, 34(1), 89–126. https://doi.org/10.1177/0149206307308913
Wang, J. (2025). An AntConc-based study of linguistic features in TED speeches from the perspective of interpersonal function. Journal of Zunyi Normal University, 27(5), 73–77.
Xia, S. A., & Hafner, C. A. (2021). Engaging the online audience in the digital era: A multimodal analysis of engagement strategies in TED talk videos. Ibérica, 42, 33–58. https://doi.org/10.17398/2340-2784.42.33
Zanotto, S. E., & Aroyehun, S. (2024). Human variability vs. machine consistency: A linguistic analysis of texts generated by humans and large language models. arXiv. https://arxiv.org/abs/2412.03025
Zhang, H., Zhang, L., Nian, H., & Zhang, L. (2024). Positive discourse analysis of attitude resources in Chat GPT news discourse from the perspective of appraisal theory. Lecture Notes on Language and Literature, 7(3), 61–66.
Zhang, N., Liu, X., & Li, D. (2026). A comparative study of lexical features in ChatGPT translation and human translation: A case study of the literary work Big Breasts and Wide Hips. Foreign Language Research, 43(2), 18–25, 113.
Zhang, Y. (2024). Discourse analysis of COP28 news reports from the perspective of attitude in appraisal theory. Studies in Linguistics and Literature, 8(2), 143-152. https://doi.org/10.22158/sll.v8n2p143
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 authors
This work is licensed under a Creative Commons Attribution 4.0 International License.