Do deepfake videos undermine our epistemic trust? A thematic analysis of tweets that discuss deepfakes in the Russian invasion of Ukraine
Open accessAbstract
Deepfakes are a form of multi-modal media generated using deep-learning technology. Many academics have expressed fears that deepfakes present a severe threat to the veracity of news and political communication, and an epistemic crisis for video evidence. These commentaries have often been hypothetical, with few real-world cases of deepfake's political and epistemological harm. The Russo-Ukrainian war presents the first real-life example of deepfakes being used in warfare, with a number of incidents involving deepfakes of Russian and Ukrainian government officials being used for misinformation and entertainment. This study uses a thematic analysis on tweets relating to deepfakes and the Russo-Ukrainian war to explore how people react to deepfake content online, and to uncover evidence of previously theorised harms of deepfakes on trust. We extracted 4869 relevant tweets using the Twitter API over the first seven months of 2022. We found that much of the misinformation in our dataset came from labelling real media as deepfakes. Novel findings about deepfake scepticism emerged, including a connection between deepfakes and conspiratorial beliefs that world leaders were dead and/or replaced by deepfakes. This research has numerous implications for future research, societal media platforms, news media and governments. The lack of deepfake literacy in our dataset led to significant misunderstandings of what constitutes a deepfake, showing the need to encourage literacy in these new forms of media. However, our evidence demonstrates that efforts to raise awareness around deepfakes may undermine trust in legitimate videos. Consequentially, news media and governmental agencies need to weigh the benefits of educational deepfakes and pre-bunking against the risks of undermining truth. Similarly, news companies and media should be careful in how they label suspected deepfakes in case they cause suspicion for real media.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
This type of misinformation can be especially common during elections and intergroup conflicts such as the Russo-Ukrainian and Israel-Gaza wars.
“This type of misinformation can be especially common during elections and intergroup conflicts such as the Russo-Ukrainian and Israel-Gaza wars (17).”
From The impact of generative artificial intelligence on socioeconomic inequalities and policy making (2024) · cites this paper as [17]
Conspiracy theories and misinformation can contribute to attitude polarization and undermine trust.
“Conspiracy theories and misinformation can contribute to attitude polarization (28) and undermine trust (17).”
From The impact of generative artificial intelligence on socioeconomic inequalities and policy making (2024) · cites this paper as [17]
Simply warning people of deepfakes or including a tag clarifying whether a piece of content is AI-generated might backfire, as such tags may reduce the believability of true content as well.
“Simply warning people of deepfakes or including a tag clarifying whether a piece of content is AI-generated might backfire, as such tags may reduce the believability of true content as well (17, 33).”
From The impact of generative artificial intelligence on socioeconomic inequalities and policy making (2024) · cites this paper as [17]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Using thematic analysis in psychology
Braun, V., & Clarke, V. (2006) · Qualitative Research in Psychology · doi:10.1191/1478088706qp063oa · Cited by 194,414 · see citing works
Tong, A., Sainsbury, P., & Craig, J. (2007) · International Journal for Quality in Health Care · doi:10.1093/intqhc/mzm042 · Cited by 42,509 · see citing works
Braun, V., & Clarke, V. (2012) · APA handbook of research methods in psychology, Vol 2: Research designs: Quantitative, qualitative, neuropsychological, and biological. · doi:10.1037/13620-004 · Cited by 10,299 · see citing works
Braun, V., & Clarke, V. (2021) · Counselling and Psychotherapy Research · doi:10.1002/capr.12360 · Cited by 3,198 · cited 1× · see citing works
“After generating and refining a set of three themes, we carried out further analysis, borrowing methodologies from discourse analysis [56], which allowed us to critically appraise the social constructions behind the textual data [57].”
MethodThe Emergence of Deepfake Technology: A Review
Westerlund, M. (2019) · Technology Innovation Management Review · doi:10.22215/timreview/1282 · Cited by 1,138 · see citing works
Vaccari, C., & Chadwick, A. (2020) · Social Media + Society · doi:10.1177/2056305120903408 · Cited by 978 · cited 2× · see citing works
“Previous quantitative research has shown that educational disclaimers can to some extent mitigate the harms of pure deepfake misinformation [18], but our research indicates that educational deepfakes may still undermine truth in comparison to unedited videos.”
BackgroundDeep Fakes: A Looming Challenge for Privacy, Democracy, and National Security
Chesney, R., & Citron, D. K. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3213954 · Cited by 880 · see citing works
Identifying Sarcasm in Twitter: A Closer Look
González‐Ibáñez, R., Muresan, S., & Wacholder, N. (2011) · Cited by 628 · see citing works
“Participant” Perceptions of Twitter Research Ethics
Fiesler, C. L., & Proferes, N. (2018) · Social Media + Society · doi:10.1177/2056305118763366 · Cited by 545 · see citing works
(2004) · The SAGE Encyclopedia of Social Science Research Methods · doi:10.4135/9781412950589.n779 · Cited by 527 · see citing works
Williams, M. L., Burnap, P., & Sloan, L. (2017) · Sociology · doi:10.1177/0038038517708140 · Cited by 423 · see citing works
Developing a Method of Critical Thematic Analysis for Qualitative Communication Inquiry
Lawless, B., & Chen, Y. (2018) · Howard Journal of Communications · doi:10.1080/10646175.2018.1439423 · Cited by 323 · see citing works
The Epistemic Threat of Deepfakes
Fallis, D. (2021) · Philosophy & Technology · doi:10.1007/s13347-020-00419-2 · Cited by 266 · see citing works
Roozenbeek, J., van der Linden, S., & Nygren, T. (2020) · Harvard Kennedy School Misinformation Review · doi:10.37016//mr-2020-008 · Cited by 214 · see citing works
Anticipating and Addressing the Ethical Implications of Deepfakes in the Context of Elections
Diakopoulos, N. A., & Johnson, D. K. (2019) · New Media & Society · doi:10.1177/1461444820925811 · Cited by 210 · see citing works
Anticipating and Addressing the Ethical Implications of Deepfakes in the Context of Elections
Diakopoulos, N., & Johnson, D. (2019) · SSRN Electronic Journal · doi:10.2139/ssrn.3474183 · Cited by 194 · see citing works
Politics and porn: how news media characterizes problems presented by deepfakes
Gosse, C., & Burkell, J. (2020) · Critical Studies in Media Communication · doi:10.1080/15295036.2020.1832697 · Cited by 148 · see citing works
Effects of Disinformation Using Deepfake: The Protective Effect of Media Literacy Education
Hwang, Y., Ryu, J. Y., & Jeong, S.-H. (2021) · Cyberpsychology, Behavior, and Social Networking · doi:10.1089/cyber.2020.0174 · Cited by 148 · see citing works
Ahmed, W., Bath, P. A., & Demartini, G. (2017) · Advances in research ethics and integrity · doi:10.1108/s2398-601820180000002004 · Cited by 146 · see citing works
Visual Mis- and Disinformation, Social Media, and Democracy
Dan, V., Paris, B., Donovan, J., Hameleers, M., Roozenbeek, J., van der Linden, S., & von Sikorski, C. (2021) · Journalism & Mass Communication Quarterly · doi:10.1177/10776990211035395 · Cited by 144 · see citing works
Potter, J. (2004) · SAGE Publications Ltd eBooks · doi:10.4135/9781848608184.n27 · Cited by 132 · see citing works
lol: new language and spelling in instant messaging
Varnhagen, C. K., McFall, G. P., Pugh, N. E., Routledge, L., Sumida-MacDonald, H., & Kwong, T. E. (2009) · Reading and Writing · doi:10.1007/s11145-009-9181-y · Cited by 127 · see citing works
Ahmed, S. (2021) · Telematics and Informatics · doi:10.1016/j.tele.2020.101508 · Cited by 105 · see citing works
Navigating the maze: Deepfakes, cognitive ability, and social media news skepticism
Ahmed, S. (2021) · New Media & Society · doi:10.1177/14614448211019198 · Cited by 99 · see citing works
The Separatist War in Donbas: A Violent Break-up of Ukraine?†
Katchanovski, I. (2016) · European Politics and Society · doi:10.1080/23745118.2016.1154131 · Cited by 91 · see citing works
“All around me are synthetic faces”: The mad world of AI-generated media
Whittaker, L., Kietzmann, T. C., Kietzmann, J., & Dabirian, A. (2020) · IT Professional · doi:10.1109/mitp.2020.2985492 · Cited by 90 · see citing works
Seeing is Believing: Exploring Perceptual Differences in DeepFake Videos
Tahir, R., Batool, B., Jamshed, H., Jameel, M., Anwar, M., Ahmed, F., Zaffar, M. A., & Zaffar, M. F. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445699 · Cited by 79 · see citing works
Russia, Ukraine, and the Future World Order
Brunk, I., & Hakimi, M. (2022) · American Journal of International Law · doi:10.1017/ajil.2022.69 · Cited by 77 · see citing works
Kerner, C., & Risse, M. (2021) · Moral Philosophy and Politics · doi:10.1515/mopp-2020-0024 · Cited by 75 · see citing works
A unified account of information, misinformation, and disinformation
Søe, S. O. (2019) · Synthese · doi:10.1007/s11229-019-02444-x · Cited by 74 · see citing works
Imperial nationalism as the driver behind Russia's invasion of Ukraine
Kuzio, T. (2022) · Nations and Nationalism · doi:10.1111/nana.12875 · Cited by 55 · see citing works
Deepfake Privacy: Attitudes and RegulationPreprint
Kugler, M. B., & Pace, C. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3781968 · Cited by 48 · see citing works
Deep learning and synthetic media
Millière, R. (2022) · Synthese · doi:10.1007/s11229-022-03739-2 · Cited by 46 · see citing works
Detecting Real-Time Deep-Fake Videos Using Active Illumination
Gerstner, C. R., & Farid, H. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00015 · Cited by 45 · see citing works
DeepFakes: uncovering hardcore open source on GitHub
Winter, R., & Salter, A. (2020) · Porn Studies · doi:10.1080/23268743.2019.1642794 · Cited by 42 · see citing works
The future of online trust (and why Deepfake is advancing it)
Etienne, H. (2021) · AI and Ethics · doi:10.1007/s43681-021-00072-1 · Cited by 41 · see citing works
A Review of Deepfake Technology: An Emerging AI Threat
Sharma, M., & Kaur, M. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-981-16-5301-8_44 · Cited by 37 · see citing works
Lee, Y., Huang, K., Blom, R., Schriner, R., & Ciccarelli, C. (2021) · Cyberpsychology Behavior and Social Networking · doi:10.1089/cyber.2020.0176 · Cited by 36 · see citing works
Subjective and Objective Evaluation of Deepfake Videos
Korshunov, P., & Marcel, S. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414258 · Cited by 32 · see citing works
Kuzio, T. (2018) · Eurasian Geography and Economics · doi:10.1080/15387216.2019.1571428 · Cited by 31 · see citing works
Deepfaking Keanu: YouTube deepfakes, platform visual effects, and the complexity of reception
Bode, L. (2021) · Convergence The International Journal of Research into New Media Technologies · doi:10.1177/13548565211030454 · Cited by 29 · see citing works
Strategic Communications and Social Media in the Russia-Ukraine Conflict
Elina, L. I., & Sanda, S. (2016) · NATO science for peace and security series. D, Information and communication security · doi:10.3233/978-1-61499-699-6-86 · Cited by 23 · see citing works
Roozenbeek, J., van der Linden, S., & Nygren, T. (2020) · Apollo (University of Cambridge) · doi:10.17863/cam.48846 · Cited by 17 · see citing works
Pseudoscience and Science Fiction
May, A. F. (2016) · Science and fiction · doi:10.1007/978-3-319-42605-1 · Cited by 16 · see citing works
The reputational cost of military aggression: Evidence from the 2022 Russian invasion of Ukraine
Asadzade, P., & Izadi, R. (2022) · Research & Politics · doi:10.1177/20531680221098337 · Cited by 16 · see citing works
Wilkerson, L. (2021) · University of Missouri School of Law Scholarship Repository (University of Missouri) · Cited by 10 · see citing works
Skepticism and the Digital Information Environment
Carlson, M. (2021) · SATS · doi:10.1515/sats-2021-0008 · Cited by 9 · see citing works
Handling Internet Activism During the Russian Invasion of Ukraine: A Campus Network Perspective
Husák, M., Laštovička, M., & Plesník, T. (2022) · Digital Threats Research and Practice · doi:10.1145/3534566 · Cited by 8 · see citing works
The Emergence of Deepfakes and its Societal Implications: A Systematic Review.
Gamage, D., Sasahara, K., & Chen, J. (2021) · Tokyo Tech Research Repository (Tokyo Institute of Technology) · Cited by 7 · see citing works
Cherviatsova, A. (2021) · Central European University Press eBooks · doi:10.7829/j.ctv26jp68t · Cited by 6 · see citing works
Can you believe your eyes? [Issues of Deepfakes]
Lamb, H. (2020) · Engineering & Technology · doi:10.1049/et.2020.0601 · Cited by 2 · see citing works
Ebim, M. A. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4056555 · Cited by 1 · see citing works
Flagged first, then by citations
Research works printed but not linked: 6
- [6]Celli F. Deepfakes Are Coming: Does Australia Come Prepared? Student Section. Canberra Law Rev. 2020; 17: 193– 205.
- [7]Sylvester S. Don’t Let Them Fake You out: How Artificially Mastered Videos Are Becoming the Newest Threat in the Disinformation War and What Social Media Platforms Should Do about It. Fed Commun Law J. 2020; 72: 369– 392.
- [10]Langa J. DEEPFAKES, REAL CONSEQUENCES: CRAFTING LEGISLATION TO COMBAT THREATS POSED BY DEEPFAKES. BOSTON Univ LAW Rev. 2021; 101: 761– 801.
- [19]Chesney R, Citron D. Deepfakes and the New Disinformation War The Coming Age of Post-Truth Geopolitics. FOREIGN Aff. 2019; 98: 147– 155.
- [25]ULLRICH QJ. Is This Video Real? The Principal Mischief of Deepfakes and How the Lanham Act Can Address It. Columbia J Law Soc Probl. 2021; 55: 1– 56.
- [66]Gamage D, Ghasiya P, Bonagiri V, et al. Are Deepfakes Concerning? Analyzing Conversations of Deepfakes on Reddit and Exploring Societal Implications. CHI Conf Hum Factors Comput Syst [Internet]. New Orleans LA USA: ACM; 2022 [cited 2022 May 4]. p. 1– 19.
Cited by
Sorted by citations
The impact of generative artificial intelligence on socioeconomic inequalities and policy making
Capraro, V., Lentsch, A., Acemoglu, D., Akgun, S., Akhmedova, A., Bilancini, E., Bonnefon, J.-F., Brañas-Garza, P., Butera, L., Douglas, K. M., et al. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae191 · Cited by 371 · see citing works
Human performance in detecting deepfakes: A systematic review and meta-analysis of 56 papers
Diel, A., Lalgi, T., Schröter, I. C., MacDorman, K. F., Teufel, M., & Bäuerle, A. (2024) · Computers in Human Behavior Reports · doi:10.1016/j.chbr.2024.100538 · Cited by 64 · see citing works
Lin, L. S. F. (2025) · Frontiers in Law · doi:10.6000/2817-2302.2025.04.02 · Cited by 9 · see citing works
Görgen, A., & Saygıner, C. (2025) · Güvenlik Stratejileri Dergisi · doi:10.17752/guvenlikstrtj.1706990 · Cited by 1 · see citing works
- Authors
- Twomey, J., Ching, D., Aylett, M. P., Quayle, M., Linehan, C., & Murphy, G.
- Affiliations
- Science Foundation Ireland (IE)
- Heriot-Watt University (GB)
- Corresponding author
- Twomey, John
- Venue
- PLOS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2023-10-25
- Volume
- 18
- Issue
- 10
- Pages
- e0291668
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC10599512
- OpenAlex
- W4387950028
- Cited by
- 87
- References
- 52
- Citations by year
- 2026: 26 · 2025: 36 · 2024: 23 · 2023: 2
- FWCI
- 37.31
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLoS ONE (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Figshare (submittedVersion, cc-by-nc-sa)
- University of Limerick Institutional Repository (University of Limerick) (submittedVersion, cc-by-nc-sa)
- University of Limerick Institutional Repository (University of Limerick)
- Funders
- LERO/SFI (PP5004/13/RC/2094_2)
- Topics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Social Sciences › Political Science and International Relations › European and Russian Geopolitical Military Strategies
- Keywords
- deepfakes, epistemic trust, Russian invasion of Ukraine, misinformation, conspiracy beliefs, prebunking, deepfake detection, video evidence, Twitter, political communication
- MeSH terms
- Affect, Communication, Educational Status, Humans, Ukraine, Trust, Social Media
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
