Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Law of One's Own: The Inefficacy of the DMCA for Non-Consensual Intimate Media

Li Qiwei, Shihui Zhang, Samantha Paige Pratt, Andrew Timothy Kasper, Eric Gilbert, Sarita Schoenebeck

TL;DR

Non-consensual intimate media (NCIM) causes severe personal and societal harm, yet the DMCA—designed to protect copyright—may be ill-suited for rapid NCIM removal. The authors analyze a large Lumen-based dataset (over 54,000 NCIM-related DMCA tickets and 85 million URLs) and apply survival analysis to measure time-to-takedown on hosting platforms and time-to-deindexing on Google, distinguishing commercial from non-commercial NCIM. They find that host removals are slow (median times on the order of weeks to months for non-commercial NCIM) and that Google deindexing occurs faster but does not remove content from hosts, with larger delays for commercial NCIM. The study argues for NCIM-specific legislation that enforces rapid takedown across platforms (potentially overriding certain Section 230 protections) and calls for policy design that balances due process with swift removal, especially to protect victim-survivors and curb deepfake-related harms.

Abstract

Non-consensual intimate media (NCIM) presents internet-scale harm to individuals who are depicted. One of the most powerful tools for requesting its removal is the Digital Millennium Copyright Act (DMCA). However, the DMCA was designed to protect copyright holders rather than to address the problem of NCIM. Using a dataset of more than 54,000 DMCA reports and over 85 million infringing URLs spanning over a decade, this paper evaluates the efficacy of the DMCA for NCIM takedown. Results show that for non-commercial requests, while more than half of URLs are deindexed from Google Search within 48 hours, the actual removal of content from website hosts is much slower. The median infringing URL takes more than 45 days to be removed from website hosts, and only 5.39% URLs are removed within the first 48 hours. Additionally, the most frequently reported domains for non-commercial NCIM are smaller websites, not large platforms. We stress the need for new laws that ensure a shorter time to takedown that are enforceable across big and small platforms alike.

A Law of One's Own: The Inefficacy of the DMCA for Non-Consensual Intimate Media

TL;DR

Non-consensual intimate media (NCIM) causes severe personal and societal harm, yet the DMCA—designed to protect copyright—may be ill-suited for rapid NCIM removal. The authors analyze a large Lumen-based dataset (over 54,000 NCIM-related DMCA tickets and 85 million URLs) and apply survival analysis to measure time-to-takedown on hosting platforms and time-to-deindexing on Google, distinguishing commercial from non-commercial NCIM. They find that host removals are slow (median times on the order of weeks to months for non-commercial NCIM) and that Google deindexing occurs faster but does not remove content from hosts, with larger delays for commercial NCIM. The study argues for NCIM-specific legislation that enforces rapid takedown across platforms (potentially overriding certain Section 230 protections) and calls for policy design that balances due process with swift removal, especially to protect victim-survivors and curb deepfake-related harms.

Abstract

Non-consensual intimate media (NCIM) presents internet-scale harm to individuals who are depicted. One of the most powerful tools for requesting its removal is the Digital Millennium Copyright Act (DMCA). However, the DMCA was designed to protect copyright holders rather than to address the problem of NCIM. Using a dataset of more than 54,000 DMCA reports and over 85 million infringing URLs spanning over a decade, this paper evaluates the efficacy of the DMCA for NCIM takedown. Results show that for non-commercial requests, while more than half of URLs are deindexed from Google Search within 48 hours, the actual removal of content from website hosts is much slower. The median infringing URL takes more than 45 days to be removed from website hosts, and only 5.39% URLs are removed within the first 48 hours. Additionally, the most frequently reported domains for non-commercial NCIM are smaller websites, not large platforms. We stress the need for new laws that ensure a shorter time to takedown that are enforceable across big and small platforms alike.
Paper Structure (26 sections, 17 figures, 3 tables, 2 algorithms)

This paper contains 26 sections, 17 figures, 3 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Background
  3. Non-consensual intimate media
  4. Intellectual property
  5. Using intellectual property law for NCIM
  6. Method
  7. Research questions
  8. Sending and collecting DMCA tickets
  9. Filtering and classification
  10. Collecting URL up time
  11. Collecting Google Search deindex time
  12. Survival analysis
  13. Findings
  14. RQ 1: Platform hosts removal time
  15. RQ 2: Google Search deindex time
  16. ...and 11 more sections

Figures (17)

  • Figure 1: Method---Data collection, monitoring, and analysis. We query Lumen API for relevant NCIM tickets, which are then processed by a Python script that extracts infringing URLs and monitors them over a span of 4 weeks. The URLs are tracked for HTTP status via curl to the web page and Google Search indexing via the Google Search API. We use this data to answer: RQ 1: time-to-takedown (TTT) from web hosts RQ 2: TTT from Google Search RQ 3: changes and characteristics of NCIM reports and web hosts, 2012-2024.
  • Figure 2: Example of a ticket filed on Lumen database, with identifying information removed. Each ticket contains sender information, which could be a legal or otherwise authorized representative of the copyright holder. The recipient of the ticket is an online platform, such as Google. The platform then submits these tickets to Lumen. Tickets contain a description of the nature of the infringement, which we use the description of the ticket for subsequent qualitative analysis. Tickets include a URL to the original content, and lists URLs which contain the copyrighted content.
  • Figure 3: Timeline of data collection. The top row shows browser requests using curl to check if the infringing URL is still active. The bottom row represents the tickets we collected, some of which were filed before our data collection began. We collected tickets dating back 4 weeks prior to the start of the study and continued to monitor and gather new tickets daily for the next 4 weeks. Due to this left-truncated collection, the oldest ticket in the set is 8 weeks old. As a result, we monitored potentially active URLs for up to 8 weeks after their submission.
  • Figure 4: HTTP Status Monitoring
  • Figure 5: Proportion of infringing NCIM URLs active on web hosts, plotted against days since reported. Highlight shows only 4.02% of all URLs are removed after the critical first 48 hours after reporting. More than half---52.66%---of all URLs are not removed within the 61-day observation period.
  • ...and 12 more figures