Abstract
Materials and methods. The article presents a theoretical and empirical study based on scholarly literature on digital trust, digital identity, media literacy, AI literacy and synthetic media, as well as a questionnaire survey of students (N414). The survey identified academic tasks in which recognition and verification of AI-generated content, source checking and clarification of authorship boundaries become relevant.
Results. Digital authenticity is interpreted not as a merely technical property of content but as a socially constructed quality formed at the intersection of authorship, source reputation, platform infrastructure, user experience, academic integrity and institutional expectations of transparency. The survey shows that GenAI is already included in everyday academic tasks, while source verification and voluntary disclosure of AI assistance remain unstable practices. A five-level model of AI-content verification is proposed: formal, contextual, source-based, institutional and collective-expert levels.
Discussion. It is argued that universities should move beyond a purely detector-based logic toward an educational model focused on AI literacy, transparent rules and responsible authorship.
Keywords
- digital authenticity
- generative artificial intelligence
- AI-generated content
- digital trust
- university educational environment
- students
- academic integrity
- AI literacy
- media literacy
- content verification
References
Kreps, S., McCain, R.M., Brundage, M. All the News That’s Fit to Fabricate: AI-Generated Text as a Tool of Media Misinformation. Journal of Experimental Political Science. 2022; 9 (1): 104–117. https://doi.org/10.1017/XPS.2020.37
VKontakte study: every fourth Runet user is loyal to AI content. CNews. 2025, July 28. https://www.cnews.ru/news/line/2025-07-28_issledovanie_vkontakte (accessed on: 20.05.2026). (In Rus.)
Davydov, S.G., Matveeva, N.N., Saponova, A.V. Deepfakes as a social threat. Vestnik of Moscow University. Series: Journalism. 2025; (5): 54–78. https://doi.org/10.55959/msu.vestnik.journ.5.2025.5478. (In Rus.)
Trust in AI: analytical review. VCIOM. 2024, Dec. 24. URL: https://wciom.ru/analytical-reviews/analiticheskii-obzor/doverie-k-ii (accessed on: 20.05.2026). (In Rus.)
AI: your new best friend? VCIOM. 2024, Dec. 3. URL: https://wciom.ru/analytical-reviews/analiticheskii-obzor/ii-vash-novyi-luchshii-drug (accessed: 20.05.2026). (In Rus.)
Jisc. Student perceptions of AI 2025. 2025, May 22. URL: https://www.jisc.ac.uk/reports/student-perceptions-of-ai-2025 (accessed on: 20.05.2026).
Kuzminov, Ya.I., Kruchinskaya, E.V., Gruzdev, I.A., Naumov, A.A. Falling behind and getting ahead: student use of generative AI in education. Higher Education in Russia. 2025; 34 (6): 9–35. https://doi.org/10.31992/0869-3617-2025-34-6-9-35. (In Rus.)
Wang, F., Li, N., Cheung, A.C.K., Wong, G.K.W. In GenAI we trust: an investigation of university students' reliance on and resistance to generative AI in language learning. International Journal of Educational Technology in Higher Education. 2025; 22:59. https://doi.org/10.1186/s41239-025-00547-9
Dialog Regions detected more than 600 socially dangerous deepfakes in 2025. TASS. 2026, Apr. 13. URL: https://tass.ru/obschestvo/27089809 (accessed on: 20.05.2026). (In Rus.)
Liang, W., Yuksekgonul, M., Mao, Y., Wu, E., Zou, J. GPT detectors are biased against non-native English writers. Patterns. 2023; 4 (7):100779. https://doi.org/10.1016/j.patter.2023.100779
Labeling AI content in Russia: legislative framework and requirements. Sber Developers. 2026, Apr. 17. URL: https://developers.sber.ru/help/business-development/labeling-of-ai-content (accessed on: 20.05.2026). (In Rus.)