What the TTP Investigation Revealed About Meta and GatherOne
Facebook and Instagram ran thousands of advertisements for AI-powered “nudify” apps – tools that strip clothing from photos to create non-consensual explicit imagery – despite the platforms’ public ban on such content. A probe by the Tech Transparency Project (TTP) found that between April and June 2026, Beijing-based GatherOne Inc., one of Meta’s few official advertising partners in China, was responsible for approximately 7,600 such ads.
The report singled out an app called BAfter, created in China, which included a “ShareZone” that hosted pornographic face-swap images; reviewers on its Google Play store page flagged the presence of child sexual abuse material. TTP also discovered that a cluster of these ads directed users to the app, which was listed as suitable for all ages. Meta said the ads were eventually removed and that it has since banned BAfter and links to several other nudify apps, while GatherOne claimed it suspended new accounts for AI nudifying and face-swap services and is undertaking a compliance overhaul.
The report lands amid heightened scrutiny of deepfake porn. The US ‘Take It Down Act,’ which criminalises the publication of non-consensual sexual content and forces platforms to remove it, took effect last year, and the Federal Trade Commission began enforcement in May. For Meta, the findings also cast a harsh light on its China advertising business, which contributed $18.4 billion – 11% of global revenue – in 2024, much of it funnelled through authorised resellers like GatherOne.
Why Meta’s Ad Enforcement Failed and What’s at Stake
How an Authorised Reseller Slipped Through the Cracks
GatherOne is one of just 11 “official agents” Meta lists on its Chinese business site, giving it privileged access to place ads for local companies that cannot directly use Facebook or Instagram inside China. The firm handles gaming, e-commerce, finance and entertainment clients across Meta, Google, TikTok and X. According to TTP, many of the nudify ads were clearly suggestive – images of young women in sexual poses with text like “Create a video with any face” – yet they survived initial review. While Meta stresses that all advertisers must comply with its standards and that non-compliance can trigger ad cancellations, financial penalties or account termination, the sheer volume of 7,600 ads over three months suggests a systematic failure, not a one-off mistake.
The Bigger Picture: AI-Fuelled Abuse and Teen Exposure
The scandal highlights how cheap, accessible generative AI has supercharged the creation and spread of deepfake pornography. A George Mason University survey of US teenagers found more than half had used nudification tools on themselves or others, and over a third had been victims. Researcher Chad M.S. Steel noted that many teens already know how to access these services via messaging apps or direct websites, meaning platform-level bans alone may have limited impact. Nevertheless, when an authorised advertising pipeline funnelled millions of downloads to these tools – the apps studied had been downloaded more than 2.5 million times globally, according to AppMagic – it raises serious questions about whether Meta’s commercial incentives are aligned with user safety.
Regulatory and Reputational Fallout
Meta is already on the back foot. A former executive turned whistleblower testified that the company offered censorship tools to China and helped it advance in the AI arms race, while internal documents reported by Reuters indicated that about 19% of Meta’s China revenue in 2024 came from ads for banned content such as porn, scams and illegal gambling. With the FTC now actively enforcing the Take It Down Act, this latest report offers a fresh, detailed case study that lawmakers and regulators can use to press for tighter obligations on ad networks and app stores. For Meta, the financial equation is stark: its lucrative China ad feed – more than double the 2022 level – relies on a model that critics say treats policy violations as a cost of doing business.
Next Steps for Platforms, Advertisers, and Victims
- Expect Meta to tighten its ad review process for partners in restricted markets, potentially introducing real-time AI screening for nudify-specific language and imagery, while facing pressure to disclose more about its authorised reseller network.
- Advertisers using Chinese intermediaries should brace for more intrusive compliance checks and possible account freezes as Meta seeks to demonstrate it is policing its supply chain.
- The FTC is likely to reference this report in ongoing discussions about platform accountability; any enforcement action could set precedents for holding ad networks strictly liable for the content they distribute.
- Victims and advocacy groups may use the evidence to demand faster, more transparent takedown mechanisms, and lawyers could cite the report in lawsuits arguing that platforms profited from non-consensual imagery.
- Brand-safety technology vendors and compliance consultancies could see increased demand as platforms and agencies scramble to audit their ad inventory for deepfake-related risks.
Risk & Opportunity Assessment
| Commercial Risk | High | Meta’s $18.4 billion China ad revenue (2024) is heavily dependent on a small group of authorised resellers like GatherOne. Any disruption to that model – whether from regulatory crackdowns, advertiser boycotts or investor lawsuits – could hit a growth engine that has more than doubled in two years. |
| Competitive Risk | Medium | While all major platforms face similar deepfake-ad problems, Meta’s heavy reliance on China-based resellers makes it uniquely vulnerable. Competitors that can demonstrate tighter controls may win brand-safety conscious advertisers. |
| Regulatory Risk | High | The FTC is actively enforcing the Take It Down Act against non-consensual intimate imagery, and this report provides a concrete example of ad-network failings. It could fuel new legislation mandating immediate removal and stricter vetting of international ad partners. |
| Reputation Risk | Critical | Ads promoting tools linked to child sexual abuse material appeared alongside legitimate content on Meta’s flagship platforms. Combined with prior whistleblower testimony and internal leaks about banned-content revenues, the scandal deepens a narrative that Meta prioritises ad dollars over user safety. |
| Technology Disruption | Low | The underlying generative AI technology is not a direct threat to Meta’s business model, but its inability to police AI-driven abuse may accelerate calls for platform-level algorithmic liability, potentially reshaping how all user-generated content is treated under the law. |
| Commercial Opportunity | Medium | The crisis could spur spending on AI-based content moderation tools and third-party ad auditing services, benefiting compliance-tech firms. It may also push responsible advertisers toward verified ad-exchanges, creating a premium segment for platforms that prove their supply chains are clean. |
Comments 0