# [IGF 2026 WS #31 Press Delete: Scaling Global Protection from Image Abuse](https://intgovforum.org/en/content/igf-2026-ws-31-press-delete-scaling-global-protection-from-image-abuse) ## Description Non-consensual intimate image (NCII) abuse, the sharing or threatened sharing of private sexual images without consent, has become one of the most direct harms the Internet inflicts on women and girls. AI has made it dramatically worse: "nudify" tools now generate synthetic intimate images of real people in seconds, at scale, from nothing but a clothed photograph. NCII abuse now sits squarely inside the IGF 2026 theme of governing the Internet in the age of intelligence, because the same AI capabilities that create the harm can also be used to detect and prevent it. One detection and removal response works. Hashing technology stops the same intimate image reappearing across major platforms within seconds, free of charge, without the image ever leaving a survivor's device. Years of deployment have generated over two million protected images and removal rates above 90% where the tool is used. Yet most victims worldwide have never heard of it, most platforms have not integrated it, and most support organisations in the Global South have no way to access it. The gap is not technical. It is infrastructural, geographic, and political. This session asks what it takes to turn a working solution into genuinely global infrastructure, without recreating the exclusions that left most of the world unprotected in the first place. Drawing on direct frontline experience from the UK helpline that pioneered hash-based prevention, the largest global platform deploying it at scale, and a leading digital rights organisation in South Asia, speakers will debate who pays, who trains, who is accountable when a platform declines to participate, and what AI-generated imagery means for hash-based approaches designed around authentic images. The audience will work through a live scaling dilemma in small groups before reporting back, producing shared, contestable conclusions rather than a single institutional narrative. ## Policy Question(s) 1. What stops a survivor in Karachi, Lagos, or Manila from accessing the same image-removal protection available in London within minutes of a report, and whose responsibility is it to close that gap? 2. What does the rise of AI-generated intimate imagery mean for hash-based detection approaches built around authentic images, and does it require a fundamentally different technical or governance response? 3. Who should bear the cost, liability, and accountability when a platform declines to integrate cross-border image-matching infrastructure, and what recourse do survivors have in that gap? ## Expected Outcomes Participants will leave with a concrete, comparative picture of what scaling image-based abuse prevention globally actually requires, including the specific platform, NGO-capacity, and AI-related barriers that frontline organisations are encountering now. The session will produce a short set of scaling principles drawn from the live group exercise, capturing where participants converged and where genuine disagreement remained on cost, liability, platform accountability, and AI-generated imagery. These will be shared with the multi-country partner NGO network connected to this work and published publicly as a practical reference for NGOs, regulators, and platforms considering cross-border infrastructure. Speakers will also explicitly identify where AI-generated imagery creates gaps in existing detection approaches that the wider IGF community and standards bodies need to address, connecting this session's findings to the IGF 2026 theme of governing the Internet in the age of intelligence. ## Reference Document Reference Document ## Format Classroom **Duration (minutes):** 90 **Format description:** A 90-minute classroom-format session with a structured breakout segment is best suited to this topic because the core value lies in surfacing genuine disagreement, not in presenting a settled position. The first half uses short, focused speaker contributions and moderator-led exchange to establish the real barriers to scaling image-based abuse prevention across borders. The second half moves participants into small facilitated groups, each working through a live scaling dilemma drawn from the panel's own frontline experience, before reporting conclusions back to the room. This format was used successfully in our IGF 2024 session and produced substantially richer audience contributions than a panel-only format. Ninety minutes allows enough time for both the comparative speaker discussion and a genuine breakout exercise without compressing either into a token gesture, and classroom seating allows online and onsite participants to be grouped into mixed discussion clusters rather than treated as two separate audiences. ## Hybrid Format **Hybrid Format:** Online and onsite participants will be merged into the same small breakout groups for the live scaling exercise, using a shared online document so every group, regardless of location, contributes to the same visible output. A dedicated online moderator will manage remote group participation, relay contributions into the room during report-back, and ensure online attendees speak before the floor closes. Speaker contributions will include both onsite and remote participants so no single stakeholder perspective is confined to one mode of participation. We will use a shared collaborative document for breakout notes, visible to the whole room and online audience in real time, and a brief live poll at the session's midpoint to surface initial audience views before the breakout discussion begins. This structure was tested successfully in our previous IGF sessions, where online moderators represented remote participants' interventions directly into the discussion. ## Organizers **Organizer 1:** David Wright, **Organizer 2:** Boris Radanovic, 🔒South West Grid for Learning ## Speakers **Speaker 1:** Sophie Mortimer, Civil Society, Western European and Others Group (WEOG) **Speaker 2:** Nighat Dad, Civil Society, Asia-Pacific Group **Speaker 3:** Cindy Southworth, Private Sector, Western European and Others Group (WEOG) **Speaker 4:** Victor Samuel, Civil Society, African Group ## Moderator David Wright, Civil Society, Western European and Others Group (WEOG) ## Online Moderator Boris Radanovic, Civil Society, Western European and Others Group (WEOG) ## Rapporteur David Wright, Civil Society, Western European and Others Group (WEOG) ## Participant Benefits What will participants gain from attending this session? ## SDGs 5. Gender Equality 5.2 5.b 16. Peace, Justice and Strong Institutions 16.10 16.3 17. Partnerships for the Goals 17.16 17.17