The DOI matches its Crossref record; 36 of the 37 references it cites were found in a registry or an index; 4 of the 114 sources its claims cite were checked for support.
Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions
Open accessAbstract
Observers fear that deepfakes will shake the very foundations of democracy. Notwithstanding, in-depth scholarly analyses of deepfakes' political impact are rare, and do not consider theories of democracy. This contribution helps close this research gap, drawing on Warren's problem-oriented democracy theory, as well as theories of deliberative democracy and contributions on the role of trust in democracies. I identify three core functions of democratic systems and their normative foundations, namely empowered inclusion, collective agenda and will formation (supported by deliberation), and collective decision-making. Based on a literature and media analysis, I systematize different types of deepfakes serving either disinformation or hate speech and outline how they weaken core democratic functions and norms: Deepfakes impede citizens' empowered inclusion in debates and decisions that affect them, e.g. by hampering efforts to hold political representatives accountable or further marginalizing certain societal groups such as women or ethnic minorities. Deepfakes also undermine collective agenda and will formation by threatening the epistemic quality of deliberation as well as citizens' mutual empathy and respect. This culminates in a decreased legitimacy of collective decisions taken, which is additionally threatened by pervasive (but mostly speculative) fears of deepfake election manipulation. My analysis has implications for (future) governance efforts addressing deepfakes. Such efforts are increasing, e.g. on the part of social media platforms, but also (supra-)national regulatory bodies.
Overview
CoChat Summary
## TL;DR
This paper argues that deepfakes threaten democracy mainly by degrading trust, deliberation, and inclusion, and it grounds that claim in Warren’s problem-based democracy theory plus deliberative democracy theory. Based on a review of more than 300 academic, media, and internet sources, it systematizes harmful deepfake uses into disinformation and hate speech and argues that their political effects are often real but frequently more speculative than decisive in elections.
## Key findings
- The paper identifies three core democratic functions threatened by deepfakes: empowered inclusion, collective agenda and will formation, and collective decision-making.
- On the basis of a review of more than 300 sources, it distinguishes six political deepfake applications/effects: election manipulation, attacking political opponents, foreign interference, the liar’s dividend, weakening news media and journalism, and deepfake hate speech.
- It argues that deepfakes can cause informational trust decay and societal trust decay, which undermine deliberation, mutual respect, and the legitimacy of collective decisions.
- The paper says election-related deepfakes have not significantly altered any democratic election to date, and that the 2022 Zelensky deepfake was the first documented attempt at wider-scale foreign political interference through deepfakes, but it was ineffective.
## Methods
- The paper is a theoretical and literature-based analysis anchored in Warren’s problem-based democracy theory and systemic deliberative democracy theory (especially Mansbridge et al.).
- Empirical material comes from more than 300 academic contributions, media reports, and internet publications in English and German, with academic sources found via Google Scholar searches from January 2020 to May 2021 and other sources via Google Alerts from June 2020 to April 2021.
- The author used an inductive systematization of deepfake use cases, then assessed them through the paper’s democratic-theory framework.
## Limitations
- The author explicitly says the classification is inductive and not the only conceivable categorization.
- Only German- and English-language sources were reviewed, and the evidence base includes many non-peer-reviewed media, think tank, company, and government reports.
- The theoretical lens draws on a limited number of democracy-theory contributions from the Global North.
- Evident from the text: because the paper relies heavily on dispersed secondary sources and many reported cases are described as speculative, rare, or not yet decisive, some conclusions—especially about election manipulation and foreign interference—remain limited by the available evidence (stated/implied in the paper).
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
4 Valid · 8 Not valid · 100 Not checked · 106 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Deepfake technology is often still unnecessary and inefficient (Meneses, 2021: 7).23”
§ Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions > Deepfake Disinformation and Democracy > Election Manipulation A widespread fear is deepfakes’ use mainly by domestic (and non-state foreign) actors to spread disinformation in the context of elections (see, e.g., Diakopoulos & Johnson, 2019; Chesney & Citron, 2019: 1774–1778).22 However, my review of scholarly literature and media reports reveals only few such deepfakes: In 2018, a Flemish socialist party spread doubt about climate change using a crude (and labelled) deepfake video of Trump (Parkin, 2019). Various parties posted manipulated images and videos during the Indian parliamentary elections 2019 (Goel & Frenkel, 2019), but none were deepfakes. In the 2020 US presidential elections, fears of deepfakes were widespread, but largely did not materialize. I found only few (potential) cases here: In March 2020, Trump himself retweeted a fabricated video of Biden pulling grimaces. However, the original post was labelled a deepfake (Frum, 2020), the video is hardly deceptive, and it has been doubted whether it is technically a deepfake (Cole, 2020). In May 2020, Trump tweeted a deepfake of himself superimposed into the film “Independence Day” (Papenfuss, 2020). While technically more clearly a deepfake, its deceptive potential is also doubtful. Similarly, “robocalls” to millions of voters in Michigan in autumn 2020 urged them not to vote or spread disinformation about the elections. However, the caller sounded “robotic” (Romm & Stanley-Becker, 2020), rather than convincingly emulating a human. Also, in October 2020, a fake intelligence document about Biden’s son and his alleged business connections with China surfaced “on the right-wing internet” and was later spread, e.g., by QAnon supporters and Republican politicians to damage Biden’s presidential candidacy (Collins & Zadrozny, 2020). The document’s main author was a fake persona with a deepfake profile picture. Interestingly, this was soon uncovered, but did not impede the document’s dissemination (ibid.). Thus, while the (wider) conspiracy theory surrounding Biden’s son may have impacted the US elections, deepfake-based deception was arguably not decisive. In my assessment, deepfakes have thus not significantly altered any democratic election to date. A major reason may be that less sophisticated “cheapfakes” or even just the simple misattribution of images and videos are still sufficient to “create [political] turmoil” (Harwell, 2019), as exemplified by cheapfakes of Biden shared in 2020 (Johnson, 2020). Deepfake technology is often still unnecessary and inefficient (Meneses, 2021: 7).23 To date, the impact of election-related deepfakes on democracy thus remains mainly speculative. They could undermine inclusion in several ways. Firstly, deepfakes may exclude citizens from debates that concern them or specific candidates if they “drown out” their voices from political fora (Tenove, 2020: 529). More directly, deepfakes that spread disinformation (or fear) about election procedures, and demeaning (e.g. pornographic) deepfakes that blackmail voters into not voting, undermine the practice of voting, which only “functions as empowerment” if it is universal (Warren, 2017: 49). Thirdly, deepfakes may challenge candidates’ fair chances in an election when used for “false claims, conspiracy theories, chauvinistic language, and imagery that stokes moral revulsion toward electoral candidates and public officials” (Tenove, 2020: 528–529). Empowered inclusion is threatened when deepfakes skew deliberation to a degree that prevents citizens from holding representatives accountable for political decisions and potential misdemeanour by voting for rivals (Warren, 2017: 48). Election-related deepfakes also threaten collective agenda and will formation: Disinformation about political candidates and programmes undermines deliberation’s epistemic quality. Respective deepfakes contribute to an environment in which the public no longer trusts what it sees and hears. This hampers or even prevents rational discussions and negotiation within a public sphere which ideally considers and reacts to real-life societal problems (Habermas, 2005: 290). Also, election-related deepfakes aim to increase polarization and decrease mutual respect, which in turn impedes deliberation. Since election-related deepfakes undermine universal empowered inclusion and deliberation’s sound epistemic basis, collective decision-making is also impeded. Arguably, election-related deepfakes have not had this detrimental effect on democracy yet—although it may well materialize in the (near) future, considering deepfakes’ growing sophistication and (cost and resource) efficiency. Importantly, however, deepfakes have caused great concern among academia, journalists, political actors, and, increasingly, the public. I argue that this fear itself—rather than deepfakes’ actual use in elections—currently constitutes electoral deepfakes’ greatest threat to democracy: It undermines citizens’ and other political stakeholders’ trust in the fairness and integrity of elections (see Diakopoulos & Johnson, 2019: 12). Thus, it fuels societal trust decay. This reduces trust in elected representatives and the legitimacy of collective decisions (see also Habermas, 2005: 288).
What the sources say
- Partly supportedcites Meneses, 2021: 7 sourceExperimental — accuracy not yet measured
Section 2.3, “Why take the risk?” (page not available in our copy), says deepfakes were sometimes unnecessary for electoral disinformation and could be costly or harder to produce than conventional fakes. That supports the point in the paper’s electoral-interference context, but not the broader claim that deepfake technology is often unnecessary and inefficient across uses. (LLM, conf=0.78)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
None of the sources cited for this claim have been checked yet.
In the paper
“Reports, e.g. by think tanks and start-ups, outline deepfakes’ political impact, again without explicit reference to democratic norms”
§ Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions > The Debate About Deepfakes and Democracy
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“in-depth analyses of deepfakes’ political impact, linked to theories of democracy, are lacking”
§ Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions > The Debate About Deepfakes and Democracy
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Accessibility has caused a massive surge in deepfakes and new applications.”
§ Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions > Abstract Observers fear that deepfakes will shake the very foundations of democracy. Notwithstanding, in-depth scholarly analyses of deepfakes’ political impact are rare, and do not consider theories of democracy. This contribution helps close this research gap, drawing on Warren’s problem-oriented democracy theory, as well as theories of deliberative democracy and contributions on the role of trust in democracies. I identify three core functions of democratic systems and their normative foundations, namely empowered inclusion, collective agenda and will formation (supported by deliberation), and collective decision-making. Based on a literature and media analysis, I systematize different types of deepfakes serving either disinformation or hate speech and outline how they weaken core democratic functions and norms: Deepfakes impede citizens’ empowered inclusion in debates and decisions that affect them, e.g. by hampering efforts to hold political representatives accountable or further marginalizing certain societal groups such as women or ethnic minorities. Deepfakes also undermine collective agenda and will formation by threatening the epistemic quality of deliberation as well as citizens’ mutual empathy and respect. This culminates in a decreased legitimacy of collective decisions taken, which is additionally threatened by pervasive (but mostly speculative) fears of deepfake election manipulation. My analysis has implications for (future) governance efforts addressing deepfakes. Such efforts are increasing, e.g. on the part of social media platforms, but also (supra-)national regulatory bodies. In 2020, deepfakes, i.e. synthetic audio-visual media of human faces, bodies, or voices, are often created using artificial intelligence (AI),1 “went mainstream” (Hao & Heaven, 2020). Besides fake porn, mundane and prosocial uses increased significantly—as did the technology’s political applications. The public debate about deepfakes centres mainly on this (malicious) political potential, as a recent discourse analysis shows (Gosse & Burkell, 2020). Policy analysts fear deepfakes’ “challenge to truth in politics” (Galston, 2020), human rights organizations believe that the technology may be exploited by authoritarian regimes (Gregory, 2021), and journalists even see democracy endangered in general (e.g. Frum, 2020; Parkin, 2019; Thomas, 2020). The threat posed by deepfakes is not entirely new. Media manipulation with both harmful and benign intent is as old as media itself. E.g. convincing image manipulation has a long trajectory for art, but also non-consensual pornography (Burkell & Gosse, 2019) and disinformation, i.e. “false, inaccurate, or mis-leading information designed, presented and promoted to intentionally cause public harm or for profit” (HLEG, 2018: 10).2 In this sense, deepfakes are just a new means of media manipulation. However, the technology is developing in an information environment already challenged by the way (dis)information is shared and spread via social media (Schick, 2020). Deepfakes’ characteristics such as their increasing quality and persuasiveness also amplify challenges posed by manipulated media: In the past, it was difficult to convincingly manipulate audio and video material, increasing trust in such media. Also, audio and video “appeal[s] at a visceral level more than any text or picture ever will” and is thus often perceived as particularly credible (Kwok & Koh, 2020: 1; see also Kietzmann et al., 2020: 136). AI has now greatly simplified and improved the manipulation and synthesis of audio and video (as well as images). This is beneficial, e.g., for commercial or artistic uses of deepfake technology, but has also increased their deceptive potential. Accordingly, in a rare experimental study on deepfakes’ effect on political attitudes, less than fifteen percent of participants doubted the authenticity of a deepfake video of a politician shown to them (Dobber et al., 2020: 78). Deepfakes’ increasing quality also means that many can no longer be uncovered without technical support, aggravating harmful deepfakes’ governance. When perfected, synthetic media will no longer be based on discernible authentic raw material at all (Diresta, 2020), making them even harder to detect—and easy to generate once the initial model is completed. Besides their realism, deepfakes’ accessibility raises concerns: Unlike earlier forms of sophisticated media manipulation, deepfake generation is increasingly commercialized and accessible even to laypersons (Ajder et al., 2019: 5). Accessibility has caused a massive surge in deepfakes and new applications. This includes benign uses, but, e.g., also deepfake revenge porn, posing new ethical and societal challenges. Deepfakes thus greatly exacerbate existing (political) concerns associated with manipulated media, and technological progress suggests this trend will only continue. It is thus vital to assess deepfakes’ societal and political impact in detail. However, few scholars and policy analysts have done so to date. Social sciences and humanities research on deepfakes is rare overall and does not relate deepfakes’ impact to core democratic norms. Greater theoretical grounding and conceptual clarity are needed to help specify and evaluate the harm (and potential) of current and future deepfake applications. This will also provide a sounder basis for societal and political responses to deepfakes. As Tenove (2020: 520) argues concerning disinformation, varying understandings of threats posed to democracy “can be used to design and justify quite different policies”. These policies in turn (dis)advantage different actors and may themselves threaten core democratic norms. It is thus crucial to specify the democratic goods threatened by deepfakes to craft adequate policy responses and protect democracy. This contribution attempts to answer the following research questions: (How) Do deepfakes threaten democracy? Which underlying normative goods or values do they affect?3 Thereby, deepfakes’ impact depends heavily on their context of use and the intentions behind it.4 Many deepfakes, e.g. in the areas of political activism, art, and education, may even strengthen democratic debate and institutions (see Pawelec, 2022). However, this paper focuses explicitly on deepfakes’ harmful potential for democracy as this lies at the centre of debate about deepfakes but nonetheless remains understudied. To answer the paper’s research questions, I draw on systemic democracy theory that specifies the functions or normative goods a system must realize to be considered democratic. I integrate Warren’s (2017) “problem-based approach” to democracy theory and deliberative democracy theory, in particular the systemic approach by Mansbridge et al. (2012). I identify three core democratic functions and their normative foundations: democratic systems must enable empowered inclusion, collective agenda and will formation, and collective decision-making.5 Deliberation, in this view, is a social practice that mainly supports collective agenda and will formation and entails certain epistemic and ethical benefits. Within this framework, trust is also essential to ensure citizens’ participation in deliberation (allowing for empowered inclusion and the shaping of collective agendas) and to enable efficient political decision-making. Based on this theoretical framework, I assess claims about different deepfakes’ harmful political impact. I augment this with a technology ethical analysis of recent deepfake use cases. Thereby, my contribution assesses a large body of literature on deepfakes’ impact, which is dispersed, often cursory, and lacking an ethical or political science focus. Empirically, the analysis is based on over 300 academic contributions, media reports, and internet publications. It is structured according to an inductive systematization of six different deepfake applications and types of impact in the political context: Three deepfake uses spread disinformation, serving differing political goals (deepfakes for election manipulation, targeted attacks on political opponents, and foreign interference). The associated trust decay permits doubt or a denial of inconvenient facts (the “liar’s dividend”) and weakens news media. Additionally, I consider pornographic and other demeaning deepfakes without obvious political agenda.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Arguably, election-related deepfakes have not had this detrimental effect on democracy yet—although it may well materialize in the (near) future, considering deepfakes’ growing sophistication and (cost and resource) efficiency.”
§ Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions > Deepfake Disinformation and Democracy > Election Manipulation A widespread fear is deepfakes’ use mainly by domestic (and non-state foreign) actors to spread disinformation in the context of elections (see, e.g., Diakopoulos & Johnson, 2019; Chesney & Citron, 2019: 1774–1778).22 However, my review of scholarly literature and media reports reveals only few such deepfakes: In 2018, a Flemish socialist party spread doubt about climate change using a crude (and labelled) deepfake video of Trump (Parkin, 2019). Various parties posted manipulated images and videos during the Indian parliamentary elections 2019 (Goel & Frenkel, 2019), but none were deepfakes. In the 2020 US presidential elections, fears of deepfakes were widespread, but largely did not materialize. I found only few (potential) cases here: In March 2020, Trump himself retweeted a fabricated video of Biden pulling grimaces. However, the original post was labelled a deepfake (Frum, 2020), the video is hardly deceptive, and it has been doubted whether it is technically a deepfake (Cole, 2020). In May 2020, Trump tweeted a deepfake of himself superimposed into the film “Independence Day” (Papenfuss, 2020). While technically more clearly a deepfake, its deceptive potential is also doubtful. Similarly, “robocalls” to millions of voters in Michigan in autumn 2020 urged them not to vote or spread disinformation about the elections. However, the caller sounded “robotic” (Romm & Stanley-Becker, 2020), rather than convincingly emulating a human. Also, in October 2020, a fake intelligence document about Biden’s son and his alleged business connections with China surfaced “on the right-wing internet” and was later spread, e.g., by QAnon supporters and Republican politicians to damage Biden’s presidential candidacy (Collins & Zadrozny, 2020). The document’s main author was a fake persona with a deepfake profile picture. Interestingly, this was soon uncovered, but did not impede the document’s dissemination (ibid.). Thus, while the (wider) conspiracy theory surrounding Biden’s son may have impacted the US elections, deepfake-based deception was arguably not decisive. In my assessment, deepfakes have thus not significantly altered any democratic election to date. A major reason may be that less sophisticated “cheapfakes” or even just the simple misattribution of images and videos are still sufficient to “create [political] turmoil” (Harwell, 2019), as exemplified by cheapfakes of Biden shared in 2020 (Johnson, 2020). Deepfake technology is often still unnecessary and inefficient (Meneses, 2021: 7).23 To date, the impact of election-related deepfakes on democracy thus remains mainly speculative. They could undermine inclusion in several ways. Firstly, deepfakes may exclude citizens from debates that concern them or specific candidates if they “drown out” their voices from political fora (Tenove, 2020: 529). More directly, deepfakes that spread disinformation (or fear) about election procedures, and demeaning (e.g. pornographic) deepfakes that blackmail voters into not voting, undermine the practice of voting, which only “functions as empowerment” if it is universal (Warren, 2017: 49). Thirdly, deepfakes may challenge candidates’ fair chances in an election when used for “false claims, conspiracy theories, chauvinistic language, and imagery that stokes moral revulsion toward electoral candidates and public officials” (Tenove, 2020: 528–529). Empowered inclusion is threatened when deepfakes skew deliberation to a degree that prevents citizens from holding representatives accountable for political decisions and potential misdemeanour by voting for rivals (Warren, 2017: 48). Election-related deepfakes also threaten collective agenda and will formation: Disinformation about political candidates and programmes undermines deliberation’s epistemic quality. Respective deepfakes contribute to an environment in which the public no longer trusts what it sees and hears. This hampers or even prevents rational discussions and negotiation within a public sphere which ideally considers and reacts to real-life societal problems (Habermas, 2005: 290). Also, election-related deepfakes aim to increase polarization and decrease mutual respect, which in turn impedes deliberation. Since election-related deepfakes undermine universal empowered inclusion and deliberation’s sound epistemic basis, collective decision-making is also impeded. Arguably, election-related deepfakes have not had this detrimental effect on democracy yet—although it may well materialize in the (near) future, considering deepfakes’ growing sophistication and (cost and resource) efficiency. Importantly, however, deepfakes have caused great concern among academia, journalists, political actors, and, increasingly, the public. I argue that this fear itself—rather than deepfakes’ actual use in elections—currently constitutes electoral deepfakes’ greatest threat to democracy: It undermines citizens’ and other political stakeholders’ trust in the fairness and integrity of elections (see Diakopoulos & Johnson, 2019: 12). Thus, it fuels societal trust decay. This reduces trust in elected representatives and the legitimacy of collective decisions (see also Habermas, 2005: 288).
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
99 more claims
Create a free account to see all 218 claims and to verify the 4 references still open.
- Probably not a substantial claim. 61
- It may state two separate claims as one. 77
- The quote only partly states this claim. 1
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?
The cited source doesn’t support the claim as written.
In the citing paper
“Still, better regulation from institutions and enhanced cooperation by social media platforms are understood as the necessary and sufficient condition, or to the very least as the most urgent measure, to protect digital citizens and democratic institutions, with no complementary role envisaged for societal dialogue, AI education, and knowledge-sharing on online experiences (Painter, 2023;Pawelec,”
From Deepfakes, Democracy, and the Ethics of Synthetic Media: A Synthesis of the SOLARIS Project (2026) · cites this paper as Pawelec, 2022
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measured
Citations
Cites
Justice as Fairness: A Restatement
Rawls, J. (2001) · doi:10.2307/j.ctv31xf5v0 · Cited by 5,761
Rawls, J. (1958) · The Philosophical Review · doi:10.2307/2182612 · Cited by 5,565
Moral Consciousness and Communicative Action.
Weberman, D., Habermas, J., Lenhardt, C., Nicholsen, S. W., & McCarthy, T. (1992) · The Philosophical Review · doi:10.2307/2185962 · Cited by 3,054
A systemic approach to deliberative democracy
Mansbridge, J., Bohman, J., Chambers, S., Christiano, T., Fung, A., Parkinson, J., Thompson, D. F., & Warren, M. E. (2012) · Deliberative Systems · doi:10.1017/cbo9781139178914.002 · Cited by 1,083 · cited 1×
“They are “watchdogs over power, representatives of citizens and communities, knowledge translators, educators of citizens, and public advocates” (Mansbridge et al., 2012: 20).”
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
(1990) · DigitalGeorgetown (Georgetown University Library) · Cited by 542
(1999) · doi:10.1017/cbo9780511659959 · Cited by 508
Kietzmann, J., Lee, L. W., McCarthy, I. P., & Kietzmann, T. C. (2020) · Business Horizons · doi:10.1016/j.bushor.2019.11.006 · Cited by 504
A problem-based approach to democratic theory
WARREN, M. E. (2017) · American Political Science Review · doi:10.1017/s0003055416000605 · Cited by 478 · cited 10×
“…are prevented from participating in public life and contributing to political decisions that affect them (see also Jankowicz, 2021) as they are not afforded “equal protections that enable [them] to use [their] empowerments”, i.e. “equal rights to vote, speak, [and] organize” (Warren, 2017: 44).”
BackgroundInfluentialThe Epistemic Threat of Deepfakes
Fallis, D. (2021) · Philosophy & Technology · doi:10.1007/s13347-020-00419-2 · Cited by 266 · cited 1×
“Humanities or social sciences analyses are rare (Godulla et al., 2021: 82).6 Philosophers have recently explored deepfakes’ relation to authenticity (Floridi, 2018), the inherent moral wrong of deepfakes (Ruiter, 2021), and the “epistemic threat” they pose (Fallis, 2020).”
Background
Research works printed but not linked: 4
- Citron, D. (2019). Sexual privacy. The Yale Law Journal, 1870–1960.
- Habermas, J. (1999). Wahrheit und Rechtfertigung: Philosophische Aufsätze (1st ed.). Suhrkamp.
- Jaiman, A. (2020). Debating the Ethics of Deepfakes. In Mirchandani, M. (Ed.), Tackling Insurgent Ideologies in a Pandemic World, ORF and Global Policy Journal, New Delhi, pp. 75–79.
- WITNESS. (2020b). Not funny anymore: Deepfakes, manipulated media, parody and mis/disinformation: Jane Lytvynenko (Buzzfeed News), Karen Hao (MIT Tech Review) and Brandi Collins-Dexter (Color of Change) in conversation with Corin Faife (WITNESS). Deepfakery Webtalk Series, 2.
Cited by
Shoaib, M. R., Wang, Z., Ahvanooey, M. T., & Zhao, J. (2023) · 2023 International Conference on Computer and Applications (ICCA) · doi:10.1109/icca59364.2023.10401723 · Cited by 202
“The cumulative effects of unchecked deepfakes and misin-formation are the erosion of public trust [28].”
BackgroundOn the way to deep fake democracy? Deep fakes in election campaigns in 2023
Łabuz, M., & Nehring, C. (2024) · European Political Science · doi:10.1057/s41304-024-00482-9 · Cited by 131
“They have found numerous applications in various fields, which may vary from being extremely useful to morally questionable, dangerously harmful or directly illegal (Farid & Schindler 2020; Pawelec 2022).”
Artificial Intelligence and Political Deepfakes: Shaping Citizen Perceptions Through Misinformation
Momeni, M. (2024) · Journal of Creative Communications · doi:10.1177/09732586241277335 · Cited by 83
“As Pawelec (2022) argues, the fear of using deepfakes endangers the reliability of elections.”
Islam, M. B. E., Haseeb, M., Batool, H., Ahtasham, N., & Muhammad, Z. (2024) · Blockchains · doi:10.3390/blockchains2040020 · Cited by 76
Hynek, N., Gavurová, B., & Kubák, M. (2025) · Journal of Innovation & Knowledge · doi:10.1016/j.jik.2025.100782 · Cited by 83
Metadata
- Authors
- Pawelec, M.
- Affiliations
- University of Tübingen (DE)
- Corresponding author
- Pawelec, Maria
- ORCID
- Pawelec, Maria
- Venue
- Digital Society
- Published
- 2022-09
- Volume
- 1
- Issue
- 2
- Pages
- 19-19
- ISSN
- 2731-4650 (ISSN-L)
- 2731-4669
- PMC
- PMC9453721
- OpenAlex
- W4294952471
- Cited by
- 131
- References
- 35
- Citations by year
- 2026: 45 · 2025: 35 · 2024: 44 · 2023: 7
- FWCI
- 30.93
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor, 7 d)
- Locations
- Digital Society (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- Bundesministerium für Bildung und Forschung (16ITA214)
- Ministerium für Wissenschaft, Forschung und Kunst Baden-Württemberg (Kapitel 1223 Titelgruppe 94 Unterteil 0058)
- Eberhard Karls Universität Tübingen
- Topics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Communication › Social Media and Politics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Keywords
- deepfakes, democracy, disinformation, hate speech, deliberative democracy, trust in democracy, will formation, group decision making, electoral manipulation, political accountability, marginalization, deepfake regulation, synthetic 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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions. https://app.cochat.ai/papers/10.1007/s44206-022-00010-6
- The paper
- Pawelec, M. (2022). Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions. Digital Society. https://doi.org/10.1007/s44206-022-00010-6
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.