Designed to abuse? Deepfakes and the non-consensual diffusion of intimate images
Open accessThe DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index; 2 of the 22 sources its claims cite were checked for support.
Abstract
The illicit diffusion of intimate photographs or videos intended for private use is a troubling phenomenon known as the diffusion of Non-Consensual Intimate Images (NCII). Recently, it has been feared that the spread of deepfake technology, which allows users to fabricate fake intimate images or videos that are indistinguishable from genuine ones, may dramatically extend the scope of NCII. In the present essay, we counter this pessimistic view, arguing for qualified optimism instead. We hypothesize that the growing diffusion of deepfakes will end up disrupting the status that makes our visual experience of photographic images and videos epistemically and affectively special; and that once divested of this status, NCII will lose much of their allure in the eye of the perpetrators, probably resulting in diminished diffusion. We conclude by offering some caveats and drawing some implications to better understand, and ultimately better counter, this phenomenon.
Overview
CoChat Summary
TL;DR
The paper argues that deepfakes may eventually reduce the appeal of non-consensual intimate images (NCII) by eroding the special epistemic and affective status that photographs and videos currently enjoy. The authors therefore offer “qualified optimism,” while stressing that deepfakes can still cause serious harm and should not be treated as a solution.
Key findings
- The authors’ core claim is a two-premise argument: NCII’s allure and harm depend on the perceived special status of photographic images/videos, and growing awareness of deepfakes will erode that status, so “in the long run” deepfake NCII may have less allure and harm.
- They state that NCII/IBSA are widespread: a survey of 6109 respondents across Australia, New Zealand, and the UK found 1 in 3 had been victims of some sort of IBSA and 1 in 6 had perpetrated at least one.
- They report results from Flynn et al. (2021): in a reanalysis focused on deepfake or altered-image IBSA, 7.6% of their sample (n = 1466) had engaged in at least one form of IBSA involving creating, sharing, or threatening to share manipulated contents, and 14% (n = 1864) had been victims of at least one form of abuse.
- The paper argues that if photographs/videos become widely seen as potentially fabricated, perpetrators may lose the “sense of intimacy” and voyeuristic payoff they currently seek, which could reduce circulation of NCII.
Methods
- This is a conceptual/philosophical essay, not an empirical study: the authors build an argument from existing literature on photography, testimony, deepfakes, and NCII/IBSA.
- They use thought experiments and philosophical comparison, especially Walton’s contrast between how we react to a photograph versus a painting, to examine how beliefs about authenticity shape trust and intimacy.
- They support the discussion with cited secondary evidence from prior surveys and qualitative studies on IBSA, deepfakes, and digital media, but they do not present new primary data.
Limitations
- The paper explicitly says its optimistic conclusion does not warrant “naive and unconstrained optimism” or a “laissez-faire attitude” toward NCII and IBSA.
- It states that the shift to skepticism depends on awareness of deepfakes, which may spread unevenly across people and may take decades; the paper also notes that older generations may never reach the same level of disillusion as digital natives.
- The text itself makes evident that the main argument is speculative and not empirically tested here: the authors repeatedly say that whether affective reactions are belief-insensitive is “an open question” and “must be addressed empirically.”
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 1 Not valid · 15 Not checked · 36 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, it should be kept in mind that, as highlighted by several scholars, deepfake intimate images can still harm even if manifestly false .”
What the sources say
- Partly supportedcites Semenzin & Bainotti, 2020 sourceExperimental — accuracy not yet measured
The Introduction (p. 1) says non-consensual intimate-image sharing inflicts psychological harm, and Contextualizing NCII (p. 2) describes it as a technology-facilitated harm. Neither passage addresses deepfake images or whether an image can cause harm when it is manifestly false; searches for deepfakes and fabricated images found no passage making that point. (LLM, conf=0.80)
- Not checked — no full text of the sourcecites Henry et al., 2020 source
- Supportedcites Öhman, 2020 source
Öhman's essay treats Deepfakes as fabricated, 'fictional' depictions — '(fictionally) starring women who have not given their consent' and 'not unlike an artificial, or at least an augmented, fantasy' (Introduction, p. 1–2) — and still concludes they are harmful: 'even if each such case is harmless when considered in isolation, the totality amounts to something more than the sum of these individual cases' and 'the phenomenon of Deepfake Pornography would still be just as harmful... since it is the phenomenon as such' (Moving towards a solution, pp. 5–6; Conclusion, p. 7). The Introduction's thought experiment explicitly strips away harm to the depicted person's wellbeing and reputation (conditions i and ii, p. 2) yet retains the wrongness of the act, so the falsity/fabrication of the image does not cancel its harm. One caveat: the paper locates that harm at the societal level (women as a collective identity, p. 5) rather than in the individual image, and it never uses the qualifier 'manifestly' — its Deepfakes are described as 'hyper-realistic' — but the claim as phrased ('can still harm even if false') is what the source asserts. (LLM, conf=0.78)
plainstates a cause- 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
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we surmise that the increased awareness of the possibilities of deepfake technology could promote skepticism as a default attitude toward (m)any images that look like photographs and videos.”
hedgedstates a possible cause- 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 conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Now, if digital photography has ignited a shift in our attitude toward images that look like photos, deepfake technology is going to accelerate this shift by allowing cheap mass production of fabricated images that are visually indistinguishable from actual photos.”
plainstates a cause- 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 conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our follow-up to Walton's experiment is aimed at suggesting that our doxastic and affective stance toward photographic images may not be carved in stone. Rather, it is at least conceivable that it can be modified given the appropriate circumstances.”
hedgedstates a possible cause- 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 conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“And we now know why, namely because visual representations like hand-made drawings lack the special epistemic and affective power typically possessed by photographs and videos.”
boostedstates a cause- 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: unclear — It is unclear why the paper says it holds.
17 more claims
Create a free account to see all 53 claims and every source verdict.
- Probably not a substantial claim. 24
- It may state two separate claims as one. 13
Citations
Cites
Epistemic Injustice: Power and the Ethics of Knowing
Fricker, M. (2007) · doi:10.1093/acprof:oso/9780198237907.001.0001 · Cited by 9,198
Visual Pleasure and Narrative Cinema
Mulvey, L. (1975) · Screen · doi:10.1093/screen/16.3.6 · Cited by 8,680 · cited 2×
“As we learn from film studies (notably from Mulvey, 1975), themassive spread of mainstream cinema has promoted and buttressed 1 The term “toxic masculinity” usually refers to that set of harmful cultural rules about masculinity that group together a set of stereotypes concerning the domination of…”
BackgroundGenerative Adversarial Networks
Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.2661 · Cited by 4,477
Stanford Encyclopedia of Philosophy 2011: Launching a "Freemium" Model
Loy, M. (2015) · doi:10.18665/sr.22379 · Cited by 1,751
Visual Pleasure and Narrative Cinema
Mulvey, L. (1975) · Feminisms · doi:10.1007/978-1-349-14428-0_27 · Cited by 1,138
Deep 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 · cited 1×
“The concerns expressed by journalists and scholars dealing with such issues (e.g. Cole, 2017; Chesney & Citron, 2019; Gosse & Burkell, 2020; Flynn et al., 2021) offer grounds for fearing that the rapid development of deepfakes has already begun making these abuses ubiquitous, exponentially…”
BackgroundCountering Misinformation and Fake News Through Inoculation and Prebunking
Lewandowsky, S., & van der Linden, S. (2021) · European Review of Social Psychology · doi:10.1080/10463283.2021.1876983 · Cited by 773 · cited 2×
“A similar approach is being taken by scholars who, following the metaphor of vaccines, employed the inoculation of ‘inactivated’ fake news to raise subjects’ defenses against online manipulations (see Lewandowsky & van der Linden, 2021).”
BackgroundBeyond ‘Revenge Porn’: The Continuum of Image-Based Sexual Abuse
McGlynn, C., Rackley, E., & Houghton, R. (2017) · Feminist Legal Studies · doi:10.1007/s10691-017-9343-2 · Cited by 471 · cited 1×
“To emphasize the continuum with other forms of abusive behaviors, some scholars have placed NCII within the broader category of image-based sexual abuse (IBSA; McGlynn et al., 2017; Herny et al., 2020: ch. 1).”
BackgroundTransparent Pictures: On the Nature of Photographic Realism
Walton, K. L. (1984) · Critical Inquiry · doi:10.1086/448287 · Cited by 427 · cited 5×
“…and altered as not, our faith in the credibility of photography will inevitably, if slowly and painfully weaken, and one of the major differences in our conceptions of paintings and photographs 11 But see (Walton, 1984, footnote 29). could all but disappear” (2000, p. 202; but see Hopkins 2012).”
BackgroundInfluentialThe Epistemic Threat of Deepfakes
Fallis, D. (2021) · Philosophy & Technology · doi:10.1007/s13347-020-00419-2 · Cited by 266
Research works printed but not linked: 5
- Caletti, G. M. (2021). Can affirmative consent Save” revenge Porn” laws? Lessons from the italian criminalization of non-consensual pornography. Va JL & Tech, 25, 112.
- Jochelson, R., Ireland, D., & Taylor, H. (2021). Clearing your history: a review of non-consensual distribution of intimate images in Canada and future responses. UBCL Rev, 54, 763.
- Kuhn, A. (1985). The power of the image. Essays on representation and sexuality. Routledge.
- Leonard, N. (2021). Epistemological problemsof testimony. In E. Zalta (Ed.), The Stanford encyclopedia of philosophy.
- Peirce, C. S. (1984). What is a sign? Reprinted in: Peirce, C. S. (1982). The writings of Charles S. Peirce: A chronological edition. Volumes 2. Peirce edition project. Indiana University Press.
Cited by
The Impact of Artificial Intelligence on Human Sexuality: A Five-Year Literature Review 2020–2024
Döring, N., Le, T. D., Vowels, L. M., Vowels, M. J., & Marcantonio, T. L. (2024) · Current Sexual Health Reports · doi:10.1007/s11930-024-00397-y · Cited by 65
Flattery, T., & Miller, C. (2024) · Philosophy & Technology · doi:10.1007/s13347-024-00812-1 · Cited by 27
“Some authors—e.g., Viola and Voto (2023), Cavedon-Taylor (2024), and Fallis (2020)—have argued (or at least suggested) that, if deepfakes become widespread and people become widely aware of this development, it might well “[shift] our default attitude toward…photographs and videos, possibly…”
InfluentialAdetayo, A. J. (2026) · Library Hi Tech News · doi:10.1108/lhtn-01-2024-0005 · Cited by 22
Deepfakes on Demand: The rise of accessible non-consensual deepfake image generators
Hawkins, W., Mittelstadt, B., & Russell, C. (2025) · Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3715275.3732107 · Cited by 20
“In combination, the potential to generate sexual content and the possibility of creating images of identifiable individuals allows for the generation of NCII [59].”
Expropriated Minds: On Some Practical Problems of Generative AI, Beyond Our Cognitive Illusions
Paglieri, F. (2024) · Philosophy & Technology · doi:10.1007/s13347-024-00743-x · Cited by 27
“…not only rooted in the very concrete possibility that such technologies may be used to create deepfakes and share non-consensual intimate images, e.g. in the context of revenge porn (for discussion, see Viola & Voto, 2023): it applies also to apparently legitimate uses of our audiovisual materials.”
Metadata
- Authors
- Viola, M., & Voto, C.
- Affiliations
- Roma Tre University (IT)
- University of Turin (IT)
- Corresponding author
- Viola, Marco
- Venue
- Synthese
- Published
- 2023-01-13
- Volume
- 201
- Issue
- 1
- ISSN
- 0039-7857 (ISSN-L)
- 1573-0964
- OpenAlex
- W4315782642
- Cited by
- 43
- References
- 52
- Citations by year
- 2026: 18 · 2025: 8 · 2024: 13 · 2023: 4
- FWCI
- 18.56
- 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)
- Locations
- Synthese (publishedVersion, cc-by)
- Institutional Research Information System University of Turin (University of Turin) (submittedVersion, cc-by)
- Iris (Roma Tre University) (submittedVersion)
- Funders
- HORIZON EUROPE European Research Council (819649)
- Università degli Studi Roma Tre
- Topics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Arts and Humanities › History › Photography and Visual Culture
- Social Sciences › Gender Studies › Gender, Feminism, and Media
- Keywords
- deepfakes, image tampering, visual experience, privacy violations
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 and qwen3.8-omni-flash).
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: Designed to abuse? Deepfakes and the non-consensual diffusion of intimate images. https://app.cochat.ai/papers/10.1007/s11229-022-04012-2
- The paper
- Viola, M., & Voto, C. (2023). Designed to abuse? Deepfakes and the non-consensual diffusion of intimate images. Synthese. https://doi.org/10.1007/s11229-022-04012-2
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.