pith. sign in

arxiv: 2602.04759 · v2 · pith:W2TK6TXZnew · submitted 2026-02-04 · 💻 cs.CY · cs.HC

How to Stop Playing Whack-a-Mole: Mapping the Ecosystem of Technologies Facilitating AI-Generated Non-Consensual Intimate Images

Michelle L. Ding , Harini Suresh , Suresh Venkatasubramanian This is my paper

Pith reviewed 2026-05-21 13:50 UTC · model grok-4.3

classification 💻 cs.CY cs.HC
keywords AI-generated non-consensual intimate imagesAIG-NCIItechnology ecosystemtaxonomyimage-based sexual abusegenerative AItech policyintervention mapping
0
0 comments X

The pith

A taxonomy of eleven technology categories maps the full ecosystem enabling AI-generated non-consensual intimate images.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes a comprehensive map of technologies that support the creation, distribution, and monetization of AI-generated non-consensual intimate images to replace fragmented, reactive responses with coordinated ones. By synthesizing over a hundred primary sources, it identifies eleven categories that span creation through infrastructural support and visualizes how they interconnect. This shared model matters because current efforts across industry, policy, and research remain siloed and difficult to evaluate, perpetuating a cycle of addressing harms one by one. A sympathetic reader would see the taxonomy as a foundation for clearer policy, industry standards, and research that anticipates how changes in one area affect others.

Core claim

The central claim is that the first comprehensive AIG-NCII technological ecosystem can be built by mapping and taxonomizing eleven categories of technologies facilitating the creation, distribution, proliferation and discovery, infrastructural support, and monetization of AIG-NCII. The ecosystem is constructed through synthesis of over a hundred primary sources from researchers, journalists, advocates, policymakers, and technologists, then demonstrated through two walkthroughs: a case study of Grok to make sense of new harms and a mapping of U.S. federal law together with sixty-three state laws to clarify the policy landscape. The authors conclude by recommending refinement of the ecosystems

What carries the argument

The AIG-NCII technological ecosystem, a visualized map and taxonomy of eleven categories that situates individual technologies within an interconnected whole, carries the argument by supplying a consistent mental model that lets stakeholders locate their interventions relative to the larger structure.

If this is right

  • Interventions can be situated and compared within the full ecosystem rather than targeted in isolation.
  • New harms can be analyzed by locating them inside the existing category structure, as illustrated by the Grok walkthrough.
  • Policy landscapes can be mapped systematically, as shown by the alignment of federal and sixty-three state laws to the ecosystem.
  • Researchers can examine relationships between categories and the ripple effects that follow from changes in any one area.
  • Stakeholders gain shared terminology that supports coordinated prevention efforts instead of repeated reactive fixes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Mapping international laws against the same taxonomy could expose regulatory gaps that single-country analyses miss.
  • Focusing oversight on the infrastructural and monetization categories might yield higher-leverage interventions than targeting only creation tools.
  • Testing the taxonomy against the next wave of video or audio generators would show whether the eleven categories remain stable or require expansion.
  • Treating the ecosystem as a network could help model how restricting one technology category redistributes activity across others.

Load-bearing premise

The synthesis of over a hundred primary sources from researchers, journalists, advocates, policymakers, and technologists produces an accurate, complete, and unbiased taxonomy of the current AIG-NCII technological ecosystem.

What would settle it

Identification of a major technology that enables AIG-NCII yet fits none of the eleven categories, or application of the taxonomy to a fresh set of emerging tools or laws that fails to clarify relationships or policy options.

Figures

Figures reproduced from arXiv: 2602.04759 by Harini Suresh, Michelle L. Ding, Suresh Venkatasubramanian.

Figure 1
Figure 1. Figure 1: The ecosystem of technologies that facilitate AI-generated non-consensual intimate images (AIG-NCII) [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Screenshot of grok.com/plans. Accessed January 11, 2026 [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Screenshot of the payment process for SuperGrok. This is the page that appears after clicking “Upgrade [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
read the original abstract

The last decade has witnessed a rapid advancement of generative AI technology that significantly scaled the accessibility of AI-generated non-consensual intimate images (AIG-NCII), a form of image-based sexual abuse that disproportionately harms and silences women and girls. There is a patchwork of commendable efforts across industry, policy, academia, and civil society to address AIG-NCII. However, these efforts lack a shared, consistent mental model that clearly situates the technologies they target within the context of a large, interconnected, and ever-evolving technological ecosystem. As a result, interventions remain siloed and are difficult to evaluate and compare, leading to a reactive cycle of whack-a-mole. In this paper, we contribute the first comprehensive AIG-NCII technological ecosystem that maps and taxonomizes 11 categories of technologies facilitating the creation, distribution, proliferation and discovery, infrastructural support, and monetization of AIG-NCII. First, we build and visualize the ecosystem through a synthesis of over a hundred primary sources from researchers, journalists, advocates, policymakers, and technologists. Then, we conduct two detailed walkthroughs to demonstrate the usefulness of the ecosystem in 1) making sense of new AIG-NCII harms using a case study of Grok and 2) mapping a clearer tech policy landscape using U.S. federal law and 63 state laws. We conclude with a vision for future AIG-NCII research that refines the edges of the ecosystem, recommending researchers to study critical relationships between technologies and potential ripple effects from different interventions. Our goal is to produce an AIG-NCII technological ecosystem that provides a clear, shared terminology and framework for stakeholders to move into the future of AIG-NCII prevention with clarity and foresight.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript claims to deliver the first comprehensive AIG-NCII technological ecosystem by synthesizing over one hundred primary sources from researchers, journalists, advocates, policymakers, and technologists into a visualized 11-category taxonomy covering creation, distribution, proliferation and discovery, infrastructural support, and monetization. It demonstrates the framework's utility through two walkthroughs—one applying the map to a Grok case study for new harm analysis and another mapping it against U.S. federal law plus 63 state laws—and concludes with recommendations for future research on inter-technology relationships and intervention ripple effects.

Significance. If the taxonomy holds, the work would supply a shared mental model and terminology to coordinate otherwise siloed efforts across industry, policy, academia, and civil society, moving beyond reactive interventions. The two applied walkthroughs add practical value by showing how the map clarifies both emerging harms and policy landscapes. The paper earns credit for its broad sourcing across stakeholder types and for its forward-looking research agenda focused on critical relationships and potential ripple effects.

major comments (1)
  1. [Synthesis section] Synthesis section (describing construction of the ecosystem from primary sources): The central claim that the synthesis yields the 'first comprehensive' and accurate 11-category taxonomy is load-bearing for the entire contribution, yet the manuscript supplies no source selection criteria, search protocol, inclusion/exclusion rules, sampling method, or validation steps (e.g., inter-rater reliability) for deriving or confirming the categories. Without these, it is impossible to assess completeness, omitted technologies, or selection bias, directly weakening the foundation for the subsequent case studies and policy mapping.
minor comments (2)
  1. [Abstract] Abstract: The reference to '63 state laws' is presented without any indication of how the laws were identified, sampled, or coded into the ecosystem categories; adding a brief clause on this would improve transparency of the policy walkthrough.
  2. [Ecosystem visualization] Ecosystem visualization: The figure mapping the 11 categories would benefit from clearer labeling of interconnections or a supplementary table listing representative technologies per category to aid reader navigation.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback and for recognizing the paper's potential to provide a shared framework for addressing AIG-NCII. We address the single major comment below and will incorporate revisions to improve methodological transparency.

read point-by-point responses

  1. Referee: [Synthesis section] Synthesis section (describing construction of the ecosystem from primary sources): The central claim that the synthesis yields the 'first comprehensive' and accurate 11-category taxonomy is load-bearing for the entire contribution, yet the manuscript supplies no source selection criteria, search protocol, inclusion/exclusion rules, sampling method, or validation steps (e.g., inter-rater reliability) for deriving or confirming the categories. Without these, it is impossible to assess completeness, omitted technologies, or selection bias, directly weakening the foundation for the subsequent case studies and policy mapping.

    Authors: We appreciate the referee highlighting the importance of methodological clarity for the synthesis process. The manuscript describes building the ecosystem from a synthesis of over one hundred primary sources but does not provide explicit details on source selection or category derivation. We agree this limits readers' ability to evaluate completeness and potential bias. In the revised manuscript, we will add a dedicated subsection in the Synthesis section that outlines our approach: sources were identified through iterative searches across academic literature, journalism archives, advocacy reports, policy documents, and technology blogs using terms related to AI-generated intimate imagery and associated tools; inclusion focused on materials describing technologies enabling creation, distribution, proliferation, discovery, support, or monetization of AIG-NCII; categories emerged through repeated author review and grouping of recurring themes until saturation; and validation occurred via internal consensus and cross-checking against the full source set. While this was a qualitative mapping rather than a formal systematic review with quantitative reliability metrics, the added description will allow better assessment of the taxonomy. This revision will strengthen the paper without changing its core claims or structure. revision: yes

Circularity Check

0 steps flagged

Descriptive synthesis and taxonomy paper shows no circularity

full rationale

The paper constructs its 11-category AIG-NCII ecosystem map exclusively through synthesis of over one hundred external primary sources from researchers, journalists, advocates, policymakers, and technologists. No mathematical derivations, equations, predictions, fitted parameters, or self-referential definitions appear in the provided text. The central claim of a comprehensive taxonomy is presented as an output of that external synthesis rather than reducing to any internal fit, prior self-citation chain, or ansatz smuggled via citation. Because the work is purely descriptive and draws its categories directly from cited external material without load-bearing self-reference, the derivation chain is self-contained and independent of the paper's own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The contribution depends on the assumption that diverse primary sources can be synthesized into a stable, exhaustive taxonomy without significant gaps or biases in coverage of the evolving technology landscape.

axioms (1)
  • domain assumption The technological ecosystem facilitating AIG-NCII can be meaningfully divided into 11 distinct categories spanning creation through monetization.
    This division forms the core structure of the claimed contribution and is presented without detailed justification of boundaries or exhaustiveness in the abstract.
invented entities (1)
  • AIG-NCII technological ecosystem with 11 categories no independent evidence
    purpose: To provide a shared mental model for evaluating and comparing interventions across stakeholders
    The ecosystem and its categories are constructed via synthesis as a new framework for organizing previously siloed efforts.

pith-pipeline@v0.9.0 · 5872 in / 1480 out tokens · 76387 ms · 2026-05-21T13:50:33.253304+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

114 extracted references · 114 canonical work pages · 2 internal anchors

  1. [1]

    Adi Robertson. 2018. Reddit bans ‘deepfakes’ AI porn communities. https://www.theverge.com/2018/2/7/16982046/ reddit-deepfakes-ai-celebrity-face-swap-porn-community-ban

    work page 2018

  2. [2]

    Agence France-Presse. 2025. OpenAI will allow verified adults to use ChatGPT to generate erotic content. https: //www.theguardian.com/technology/2025/oct/14/openai-chatgpt-adult-erotic-content

    work page 2025

  3. [3]

    Shelby Akerley. 2021. Let’s Talk About (Fake) Sex Baby: A Deep Dive Into the Distributive Harms of Deepfake Pornography.Arizona Law Journal of Emerging Technologies4, 1 (March 2021). doi:10.2458/azlawjet.5505

    work page doi:10.2458/azlawjet.5505 2021

  4. [4]

    Alana Wise. 2025. Major deepfake porn site shuts down. https://www.npr.org/2025/05/06/nx-s1-5388422/mr- deepfakes-porn-site-ai-shut-down

    work page 2025

  5. [5]

    Alex Kim. 2019. nsfw_data_scraper. https://github.com/alex000kim/nsfw_data_scraper

    work page 2019

  6. [6]

    Alexios Mantzarlis. 2025. Taking an AI nudifier to court: A conversation with the lawyer suing ClothOff. https: //indicator.media/p/taking-an-ai-nudifier-to-court-shane-vogt-lawyer-suing-clothoff

    work page 2025

  7. [7]

    Alexios Mantzarlis and Santiago Lakatos. 2025. AI Nudifiers continue to reach millions and make millions. https: //indicator.media/p/ai-nudifiers-continue-to-reach-millions-and-make-millions

    work page 2025

  8. [8]

    American Sunlight Project. 2024. Deepfake Pornography Goes to Washingon: Measuring the Prevalence of AI Generaed NonConsensual Intimae Imagery Targeting Congress. https://static1.squarespace.com/static/ 6612cbdfd9a9ce56ef931004/t/67586997eaec5c6ae3bb5e24/1733847451191/ASP+DFP+Report.pdf

    work page arXiv 2024

  9. [9]

    Samantha Bates. 2017. Revenge Porn and Mental Health: A Qualitative Analysis of the Mental Health Effects of Revenge Porn on Female Survivors.Feminist Criminology12, 1 (Jan. 2017), 22–42. doi:10.1177/1557085116654565

    work page doi:10.1177/1557085116654565 2017

  10. [10]

    Becca Branum. 2024. NDII Victims Deserve Help. Let’s Build an Effective Takedown System. https://cdt.org/insights/ ndii-victims-deserve-help-lets-build-an-effective-takedown-system/

    work page 2024

  11. [11]

    Bellingcat’s Financial Investigations Team. 2025. Faking It: Deepfake Porn Site’s Link to Tech Companies. https: //www.bellingcat.com/news/uk-and-europe/2025/01/28/deepfake-porn-sites-link-to-tech-companies/

    work page 2025

  12. [12]

    Ben Goggin and Lora Kolodny. 2024. Canada’s child safety watchdog warns parents about Wizz, a Tinder-like app for teens. https://www.nbcnews.com/tech/tech-news/canadas-child-safety-watchdog-warns-parents-wizz-tinder- app-teens-rcna138732

    work page 2024

  13. [13]

    Abeba Birhane, Vinay Uday Prabhu, and Emmanuel Kahembwe. 2021. Multimodal datasets: misogyny, pornography, and malignant stereotypes. doi:10.48550/arXiv.2110.01963 arXiv:2110.01963 [cs]

    work page doi:10.48550/arxiv.2110.01963 2021

  14. [14]

    Abeba Birhane, Ryan Steed, Victor Ojewale, Briana Vecchione, and Inioluwa Deborah Raji. 2024. AI auditing: The Broken Bus on the Road to AI Accountability. doi:10.48550/arXiv.2401.14462 arXiv:2401.14462 [cs]

    work page doi:10.48550/arxiv.2401.14462 2024

  15. [15]

    Violation of my body:

    Natalie Grace Brigham, Miranda Wei, Tadayoshi Kohno, and Elissa M. Redmiles. 2024. "Violation of my body:" Perceptions of AI-generated non-consensual (intimate) imagery. doi:10.48550/arXiv.2406.05520 arXiv:2406.05520 [cs]

    work page doi:10.48550/arxiv.2406.05520 2024

  16. [16]

    Brittney McNamara. 2017. AI Porn Raises Issues of Consent. https://www.teenvogue.com/story/ai-porn-consent

    work page 2017

  17. [17]

    Jacquelyn Burkell and Chandell Gosse. 2019. Nothing new here: Emphasizing the social and cultural context of deepfakes.First Monday(Dec. 2019). doi:10.5210/fm.v24i12.10287

    work page doi:10.5210/fm.v24i12.10287 2019

  18. [18]

    California Legislative Information. 2024. SB-981 Sexually explicit digital images. https://leginfo.legislature.ca.gov/ faces/billTextClient.xhtml?bill_id=202320240SB981

    work page 2024

  19. [19]

    Caroline Haskins. 2026. Why Are Grok and X Still Available in App Stores? https://www.wired.com/story/x-grok- app-store-nudify-csam-apple-google-content-moderation/

    work page 2026

  20. [20]

    Cecilia D’Anastasio. 2026. Musk’s Grok AI Generated Thousands of Undressed Images Per Hour on X. https://www.bloomberg.com/news/articles/2026-01-07/musk-s-grok-ai-generated-thousands-of-undressed- images-per-hour-on-x

    work page 2026

  21. [21]

    Chiara Puglielli and Anne Craanen. 2025. The ecosystem of nonconsensual intimate deepfake tools on- line. https://www.isdglobal.org/digital_dispatches/the-ecosystem-of-nonconsensual-intimate-deepfake-tools- online/?ref=404media.co

    work page 2025

  22. [22]

    Portnoff

    Caoilte Ó Ciardha, John Buckley, and Rebecca S. Portnoff. 2025. AI Generated Child Sexual Abuse Material – What’s the Harm? doi:10.48550/arXiv.2510.02978 arXiv:2510.02978 [cs]

    work page doi:10.48550/arxiv.2510.02978 2025

  23. [23]

    Redmiles, Deepak Kumar, Allison McDonald, and Lucy Qin

    Princessa Cintaqia, Arshia Arya, Elissa M. Redmiles, Deepak Kumar, Allison McDonald, and Lucy Qin. 2025. Stop the Nonconsensual Use of Nude Images in Research. doi:10.48550/arXiv.2510.22423 arXiv:2510.22423 [cs]

    work page doi:10.48550/arxiv.2510.22423 2025

  24. [24]

    City Attorney of San Francisco. 2025. City Attorney shuts down 10 websites that create nonconsensual deep- fake pornography. https://sfcityattorney.org/city-attorney-shuts-down-10-websites-that-create-nonconsensual- deepfake-pornography/

    work page 2025

  25. [25]

    Civitai. 2025. Policy & Content Adjustments. https://civitai.com/articles/13632/policy-and-content-adjustments

    work page 2025

  26. [26]

    Civitai. 2025. Terms of Service. https://civitai.com/content/tos

    work page 2025

  27. [27]

    Craig Silverman and Alexios Mantzarlis. 2026. Briefing: Grok brings nonconsensual image abuse to the masses. https://indicator.media/p/briefing-grok-brings-nonconsensual-image-abuse-to-the-masses 16 Ding et al

    work page 2026

  28. [28]

    Evaluating Concept Filtering Defenses against Child Sexual Abuse Material Generation by Text-to-Image Models

    Ana-Maria Cretu, Klim Kireev, Amro Abdalla, Wisdom Obinna, Raphael Meier, Sarah Adel Bargal, Elissa M. Redmiles, and Carmela Troncoso. 2025. Evaluating Concept Filtering Defenses against Child Sexual Abuse Material Generation by Text-to-Image Models. doi:10.48550/arXiv.2512.05707 arXiv:2512.05707 [cs]

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2512.05707 2025

  29. [29]

    Ding and Harini Suresh

    Michelle L. Ding and Harini Suresh. 2025. The Malicious Technical Ecosystem: Exposing Limitations in Technical Gov- ernance of AI-Generated Non-Consensual Intimate Images of Adults. doi:10.48550/arXiv.2504.17663 arXiv:2504.17663 [cs]

    work page doi:10.48550/arxiv.2504.17663 2025

  30. [30]

    Jasmine Eggestein and Kenneth Knapp. 2014. Fighting Child Pornography: A Review of Legal and Technological Developments.Journal of Digital Forensics, Security and Law(2014). doi:10.15394/jdfsl.2014.1191

    work page doi:10.15394/jdfsl.2014.1191 2014

  31. [31]

    Elizabeth Laird, Maddy Dwyer, and Kristin Woelfel. 2024. In Deep Trouble Surfacing Tech-Powered Sexual Harassment in K-12 Schools. https://cdt.org/wp-content/uploads/2024/09/2024-09-26-final-Civic-Tech-Fall-Polling-research- 1.pdf

    work page 2024

  32. [32]

    Emanuel Maiberg. 2024. Instagram Advertises Nonconsensual AI Nude Apps. https://www.404media.co/instagram- advertises-nonconsensual-ai-nude-apps/

    work page 2024

  33. [33]

    Emanuel Maiberg. 2025. Civitai Ban of Real People Content Deals Major Blow to the Nonconsensual AI Porn Ecosystem. https://www.404media.co/civitai-ban-of-real-people-content-deals-major-blow-to-the-nonconsensual- ai-porn-ecosystem/

    work page 2025

  34. [34]

    Emanuel Maiberg. 2025. Elon Musk’s Grok AI Will ’Remove Her Clothes’ In Public, On X. https://www.404media. co/elon-musks-grok-ai-will-remove-her-clothes-in-public-on-x/

    work page 2025

  35. [35]

    Emanuel Maiberg. 2026. Inside the Telegram Channel Jailbreaking Grok Over and Over Again. https://www. 404media.co/inside-the-telegram-channel-jailbreaking-grok-over-and-over-again/

    work page 2026

  36. [36]

    Eric Szeto, Jordan Pearson, and Ivan Angelovski. 2025. This Canadian pharmacist is key figure behind world’s most notorious deepfake porn site. https://www.cbc.ca/news/canada/mrdeepfakes-porn-website-key-figure-1.7527626

    work page 2025

  37. [37]

    Evgeny Bazarov and Jean-Philippe Paradis. 2019. nsfw_data_source_urls. https://github.com/EBazarov/nsfw_data_ source_urls

    work page 2019

  38. [38]

    Dean Fido, Jaya Rao, and Craig A. Harper. 2022. Celebrity status, sex, and variation in psychopathy predicts judgements of and proclivity to generate and distribute deepfake pornography.Computers in Human Behavior129 (April 2022), 107141. doi:10.1016/j.chb.2021.107141

    work page doi:10.1016/j.chb.2021.107141 2022

  39. [39]

    Asher Flynn, Anastasia Powell, Adrian J Scott, and Elena Cama. 2022. Deepfakes and Digitally Altered Imagery Abuse: A Cross-Country Exploration of an Emerging form of Image-Based Sexual Abuse.The British Journal of Criminology62, 6 (Oct. 2022), 1341–1358. doi:10.1093/bjc/azab111

    work page doi:10.1093/bjc/azab111 2022

  40. [40]

    Whiting, and Kazutoshi Sasahara

    Dilrukshi Gamage, Piyush Ghasiya, Vamshi Bonagiri, Mark E. Whiting, and Kazutoshi Sasahara. 2022. Are Deepfakes Concerning? Analyzing Conversations of Deepfakes on Reddit and Exploring Societal Implications. InCHI Conference on Human Factors in Computing Systems. ACM, New Orleans LA USA, 1–19. doi:10.1145/3491102.3517446

    work page doi:10.1145/3491102.3517446 2022

  41. [41]

    Genevieve Oh. 2023. The Dataset! https://docs.google.com/spreadsheets/d/1vM-J_9NvOc4xHKbqqlZWzpQ0a0zsrDX- 7Jx-wPn1hwg/edit?usp=sharing

    work page 2023

  42. [42]

    Cassidy Gibson, Daniel Olszewski, Natalie Grace Brigham, Anna Crowder, Kevin R. B. Butler, Patrick Traynor, Elissa M. Redmiles, and Tadayoshi Kohno. 2024. Analyzing the AI Nudification Application Ecosystem. doi:10.48550/ arXiv.2411.09751 arXiv:2411.09751 [cs]

    work page arXiv 2024

  43. [43]

    Github. 2025. GitHub Synthetic Media and AI Tools. https://docs.github.com/en/site-policy/acceptable-use-policies/ github-synthetic-media-and-ai-tools

    work page 2025

  44. [44]

    Generative Adversarial Networks

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative Adversarial Networks. doi:10.48550/ARXIV.1406.2661 Version Number: 1

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1406.2661 2014

  45. [45]

    Governements of Australia, Canada, Chile, Denmark, Spain, Finland, France, Iceland, Mexico, Repub- lic of Korea, United Kingdom, and Sweden. 2025. The Global Partnership for Action on Gender- Based Online Harassment and Abuse calls for gender to be an integral part of the AI Action Summit. https://www.diplomatie.gouv.fr/en/french-foreign-policy/digital-di...

    work page 2025

  46. [46]

    Government of Canada. 2025. Research Agenda on Risks from Synthetic Content. https://ised-isde.canada.ca/site/ ised/en/canadian-artificial-intelligence-safety-institute/research-agenda-risks-synthetic-content

    work page 2025

  47. [47]

    Catherine Han, Anne Li, Deepak Kumar, and Zakir Durumeric. 2024. Characterizing the MrDeepFakes Sexual Deepfake Marketplace. doi:10.48550/ARXIV.2410.11100 Version Number: 3

    work page doi:10.48550/arxiv.2410.11100 2024

  48. [48]

    Will Hawkins, Chris Russell, and Brent Mittelstadt. 2025. Deepfakes on Demand: the rise of accessible non-consensual deepfake image generators. doi:10.48550/arXiv.2505.03859 arXiv:2505.03859 [cs]

    work page doi:10.48550/arxiv.2505.03859 2025

  49. [49]

    2019.The State of Deepfakes: Landscape, Threats, and Impact

    Henry Ajder, Giorgio Patrini, Francesco Cavalli, and Laurence Cullen. 2019.The State of Deepfakes: Landscape, Threats, and Impact. Technical Report. Deeptrace. https://regmedia.co.uk/2019/10/08/deepfake_report.pdf

    work page 2019

  50. [50]

    2023.How to implement online warnings to prevent the use of child sexual abuse material

    Charlotte Hunn and Paul Watters. 2023.How to implement online warnings to prevent the use of child sexual abuse material. Australian Institute of Criminology. doi:10.52922/ti78894 Mapping the Ecosystem of Technologies Facilitating AI-Generated Non-Consensual Intimate Images 17

    work page doi:10.52922/ti78894 2023

  51. [51]

    iperov. 2018. DeepFaceLab. https://github.com/iperov/DeepFaceLab

    work page 2018

  52. [52]

    James Vincent. 2017. AI tools will make it easy to create fake porn of just about anybody. https://www.theverge. com/2017/12/12/16766596/ai-fake-porn-celebrities-machine-learning

    work page 2017

  53. [53]

    Jason Koebler. 2026. Grok’s AI CSAM Shitshow. https://www.404media.co/groks-ai-csam-shitshow/

    work page 2026

  54. [54]

    Kummerländer, F

    Max Kamachee, Stephen Casper, Michelle L. Ding, Rui-Jie Yew, Anka Reuel, Stella Biderman, and Dylan Hadfield- Menell. 2025. Video Deepfake Abuse: How Company Choices Predictably Shape Misuse Patterns. doi:10.2139/ssrn. 5829303

    work page doi:10.2139/ssrn 2025

  55. [55]

    Deepfakes

    Kaylee Williams. 2024. US States Struggle to Define “Deepfakes” and Related Terms as Technically Complex Legislation Proliferates. https://www.techpolicy.press/us-states-struggle-to-define-deepfakes-and-related-terms-as- technically-complex-legislation-proliferates/

    work page 2024

  56. [56]

    2024.Technology-Facilitated Gender-Based Violence: An Emerging Issue in Women, Peace and Security

    Kristine Baekgaard. 2024.Technology-Facilitated Gender-Based Violence: An Emerging Issue in Women, Peace and Security. Technical Report. Georgetown Institute for Women, Peace and Security. https://giwps.georgetown.edu/ resource/technology-facilitated-gender-based-violence/

    work page 2024

  57. [57]

    Minghui Li and Yan Wan. 2023. Norms or fun? The influence of ethical concerns and perceived enjoyment on the regulation of deepfake information.Internet Research33, 5 (Nov. 2023), 1750–1773. doi:10.1108/INTR-07-2022-0561

    work page doi:10.1108/intr-07-2022-0561 2023

  58. [58]

    Sophie Maddocks. 2020. ‘A Deepfake Porn Plot Intended to Silence Me’: exploring continuities between pornographic and ‘political’ deep fakes.Porn Studies7, 4 (Oct. 2020), 415–423. doi:10.1080/23268743.2020.1757499

    work page doi:10.1080/23268743.2020.1757499 2020

  59. [59]

    Mary Anne Franks. 2025. CCRI Statement on the Passage of the TAKE IT DOWN Act (S. 146). https://cybercivilrights. org/ccri-statement-on-the-passage-of-the-take-it-down-act-s-146/

    work page 2025

  60. [60]

    Kugler and Carly Pace

    Matthew B. Kugler and Carly Pace. 2021. Deepfake Privacy: Attitudes and Regulation.Northwestern University Law Review116, 3 (2021). https://scholarlycommons.law.northwestern.edu/nulr/vol116/iss3/1

    work page 2021

  61. [61]

    Clare McGlynn, Kelly Johnson, Erika Rackley, Nicola Henry, Nicola Gavey, Asher Flynn, and Anastasia Powell. 2021. ‘It’s Torture for the Soul’: The Harms of Image-Based Sexual Abuse.Social & Legal Studies30, 4 (Aug. 2021), 541–562. doi:10.1177/0964663920947791

    work page doi:10.1177/0964663920947791 2021

  62. [62]

    Clare McGlynn and Erika Rackley. 2017. Image-Based Sexual Abuse.Oxford Journal of Legal Studies37, 3 (2017), 534–561. doi:10.1093/ojls/gqw033

    work page doi:10.1093/ojls/gqw033 2017

  63. [63]

    Clare McGlynn, Erika Rackley, and Ruth Houghton. 2017. Beyond ‘Revenge Porn’: The Continuum of Image-Based Sexual Abuse.Feminist Legal Studies25, 1 (April 2017), 25–46. doi:10.1007/s10691-017-9343-2

    work page doi:10.1007/s10691-017-9343-2 2017

  64. [64]

    Pulak Mehta, Gauri Jagatap, Kevin Gallagher, Brian Timmerman, Progga Deb, Siddharth Garg, Rachel Greenstadt, and Brendan Dolan-Gavitt. 2023. Can Deepfakes be created on a whim?. InCompanion Proceedings of the ACM Web Conference 2023. ACM, Austin TX USA, 1324–1334. doi:10.1145/3543873.3587581

    work page doi:10.1145/3543873.3587581 2023

  65. [65]

    My Image My Choice. [n. d.]. #BlockMrDeepfakes and the 9500+ websites dedicated to image-based sexual abuse. https://www.change.org/p/shut-down-mrdeepfakes-and-websites-dedicated-to-image-based-sexual-abuse

    work page

  66. [66]

    My Image My Choice. 2024. DEEPFAKE ABUSE: LANDSCAPE ANALYSIS: The Exponential Rise of Deepfake Abuse in 2023 - 2024. https://www.canva.com/design/DAGLHpt6WlY/Hfztqtw_-tKza_2l1cPNrA/view?utm_content= DAGLHpt6WlY&utm_campaign=designshare&utm_medium=link&utm_source=editor#1

    work page 2024

  67. [67]

    Nana Mgbechikwere Nwachukwu. 2026. TfGBV-Grok-NCII-Dataset. https://huggingface.co/datasets/Mtechlaw/ TfGBV-Grok-NCII-Dataset

    work page 2026

  68. [68]

    National Association of Attorneys General. 2025. To the legal representatives of Google Search, Microsoft Bing, and Yahoo! Search. https://www.naag.org/wp-content/uploads/2025/08/LTRS-Combined-Payment-and-Seach-Platform- FINAL.pdf

    work page 2025

  69. [69]

    National Association of Attorneys General. 2025. To the legal representatives of Visa, Mastercard, American Express, PayPal, Google Pay and Apple Pay. https://www.naag.org/wp-content/uploads/2025/08/LTRS-Combined-Payment- and-Seach-Platform-FINAL.pdf

    work page 2025

  70. [70]

    National Center on Sexual Exploitation. 2023. Apple App Store on 2023 Dirty Dozen List for Deceptive Rating System. https://endsexualexploitation.org/wp-content/uploads/Apple-App-Store-Notification-Letter_DDL-2023_ FINAL.pdf

    work page 2023

  71. [71]

    National Center on Sexual Explotation. 2024. The Dirty Dozen List 24. https://endsexualexploitation.org/apple/

    work page 2024

  72. [72]

    National Institute of Standards of Technology. 2024. Glossary. https://airc.nist.gov/glossary/

    work page 2024

  73. [73]

    National Network to End Domestic Violence. 2024. Tech-Facilitated Sexual Abuse. https://nnedv.org/reach- update/tech-facilitated-sexual-abuse/

    work page 2024

  74. [74]

    Nick Robins-Early. 2026. Elon Musk’s Grok AI generates images of ‘minors in minimal clothing’. https://www. theguardian.com/technology/2026/jan/02/elon-musk-grok-ai-children-photos

    work page 2026

  75. [75]

    Unlimited Realm of Exploration and Experimentation

    Gaithersburg Md Nist. 2024.Reducing Risks Posed by Synthetic Content: An Overview of Technical Approaches to Digital Content Transparency. Technical Report NIST AI NIST AI 100-4. National Institute of Standards and Technology, Gaithersburg, MD. NIST AI NIST AI 100–4 pages. doi:10.6028/NIST.AI.100-4 18 Ding et al

    work page doi:10.6028/nist.ai.100-4 2024

  76. [76]

    Dick, and Hidenori Tanaka

    Maya Okawa, Ekdeep Singh Lubana, Robert P. Dick, and Hidenori Tanaka. 2025. Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task. doi:10.48550/arXiv.2310.09336 arXiv:2310.09336 [cs]

    work page doi:10.48550/arxiv.2310.09336 2025

  77. [77]

    and Dawn Hawkins

    Patrick Trueman, Esq. and Dawn Hawkins. 2023. GitHub hosting source code for sexually exploitative technology, facilitating image-based sexual abuse (IBSA), sexual exploitation, and promoting the dangerous use of generative- AI. https://endsexualexploitation.org/wp-content/uploads/Microsoft-GitHub-Notification-Letter_DDL-2023_FINAL. pdf

    work page 2023

  78. [78]

    Ivan Perov, Daiheng Gao, Nikolay Chervoniy, Kunlin Liu, Sugasa Marangonda, Chris Umé, Dpfks, Carl Shift Facenheim, Luis RP, Jian Jiang, Sheng Zhang, Pingyu Wu, Bo Zhou, and Weiming Zhang. 2020. DeepFaceLab: Integrated, flexible and extensible face-swapping framework. doi:10.48550/ARXIV.2005.05535 Version Number: 5

    work page doi:10.48550/arxiv.2005.05535 2020

  79. [79]

    Vinay Uday Prabhu and Abeba Birhane. 2020. Large image datasets: A pyrrhic win for computer vision? doi:10. 48550/arXiv.2006.16923 arXiv:2006.16923 [cs]

    work page arXiv 2020

  80. [80]

    Public Citizen. 2025. Share Tracker: State Legislation on Intimate Deepfakes. https://www.citizen.org/article/tracker- intimate-deepfakes-state-legislation/

    work page 2025

Showing first 80 references.