Meta Ran Ads for Nudify App and AI-Generated CSAM: Policy Failures in 2026

Meta ran ads on Facebook and Instagram for an app called Kromix that promised to 'nudify' female politicians, and also allowed ads containing AI-generated child sexual abuse imagery — both in direct violation of the company's own content policies. These twin failures, reported in mid-August 2026, have intensified scrutiny on Meta's ad moderation systems and come at a time when the company faces an escalating legal crisis.

What Is the Kromix App and Why Did Meta Allow Its Ads?

The Kromix app is an AI-powered tool that lets users upload photos of real women and generate fake nude images without their consent. According to reports from Wired and Ars Technica, Meta ran advertisements for Kromix on its flagship platforms — Facebook and Instagram — that explicitly targeted female US politicians. The ads encouraged users to create deepfake nude videos of these public figures, despite Meta having clear policies forbidding non-consensual intimate imagery and sexual content in ads.

Times Now News reported that Apple investigated the app after the ads came to light and removed it from the App Store. The incident is a stark example of how generative AI tools can be weaponized for harassment, and how platform ad review systems can fail to catch policy-violating content.

Meta Also Ran Ads Containing AI-Generated Child Sexual Abuse Imagery

Even more alarming, Meta also ran ads that contained AI-generated child sexual abuse imagery, according to a separate Wired investigation. These ads depicted simulated child sexual abuse material (CSAM) and were allowed to run on Meta’s platforms, despite company policies that prohibit any form of CSAM. The use of generative AI to create such imagery represents a new frontier in online harm, and Meta’s failure to catch these ads raises questions about the effectiveness of its automated moderation tools and human review processes.

Incident	App/Tool	Target	Meta Policy Violated	Outcome
Kromix nudify ads	Kromix	Female US politicians	Non-consensual intimate imagery, sexual content in ads	Apple removed app from App Store; Meta under fire
AI-generated CSAM ads	Unnamed tool (unknown)	Child sexual abuse imagery	Prohibition of CSAM in any form	Ads ran despite policy; Wired investigation revealed

Both incidents occurred in mid-August 2026, and Meta has not publicly commented on specific ad approvals. The Firstpost report notes that the Kromix ads specifically named a prominent female US politician, adding a layer of political controversy to the story.

Why This Matters: Meta’s Ad Policy Enforcement in Question

These ad failures are not isolated. Meta has a long history of struggling to enforce its own advertising policies, particularly around sensitive content. The Kromix case shows a gap in detecting apps that promise to generate non-consensual deepfakes — a category of harm that Meta has pledged to combat. The CSAM ads demonstrate an even more severe failure, as child safety is a top regulatory priority worldwide.

The timing is especially damaging. Meta is in the midst of an escalating legal battle. CNBC reported that Meta faces what one analyst called 'astronomical' consequences in a federal trial brought by a coalition of state attorneys general. The trial centers on Meta’s alleged harms to young users, and the new ad revelations could further undermine the company’s defense. Separately, a U.S. judge ruled that Meta wiped key scam evidence in a lawsuit filed by Australian billionaire Andrew Forrest, as covered by the Australian Financial Review.

Apple’s Response: Kromix Removed From App Store

One immediate consequence of the Kromix ad reports was action by Apple. According to Times Now News, Apple pulled the Kromix app from its App Store after an investigation. Apple’s App Store review guidelines prohibit apps that facilitate non-consensual intimate content. This move is notable because Apple itself has faced scrutiny over its own app review process, but in this case it acted swiftly after press reports. Meta, by contrast, allowed the ads to run on its platforms until external investigations forced the issue.

Broader Context: Meta’s Moderation Challenges and Zuckerberg’s AI Vision

Meta’s ad review system largely relies on automated machine learning tools supplemented by human reviewers. The Kromix and CSAM ads suggest that both layers failed. Meta has invested heavily in AI-based content moderation — including its own large language models — but these incidents show that adversarial actors can still slip through, especially when they use novel AI tools to generate harmful content.

At the same time, Meta CEO Mark Zuckerberg has been making the case for unleashing superintelligent AI, as documented by The Verge in a recent interview. Critics argue that as Meta pushes toward more powerful AI systems, the company’s inability to control harmful uses of existing AI on its own platforms raises serious questions about its preparedness. The juxtaposition of Zuckerberg’s futurist rhetoric and the present-day real-world harm of the company’s ad moderation failures is striking.

What Needs to Change?

Meta faces a multi-front crisis: regulatory lawsuits, public trust erosion, and now clear evidence that its ad systems are allowing dangerous content. The Kromix ads and CSAM ads are not just PR problems — they are indicators of systemic failure. To restore credibility, Meta would need to:

• Overhaul its ad review pipeline, especially for apps that use generative AI.
• Implement real-time scanning for deepfake and CSAM indicators in image and video ads.
• Establish faster takedown processes and public transparency reports.
• Cooperate fully with regulators and plaintiffs in pending litigation.

However, given the scale of Meta’s ad inventory — billions of ads per day — fixing these gaps will be technically challenging and expensive. The company may also face legislative pressure from lawmakers who see these incidents as proof that voluntary moderation is insufficient.

Conclusion

In August 2026, Meta ran ads that promoted the non-consensual deepfake app Kromix targeting female politicians and ads that contained AI-generated child sexual abuse imagery. These events expose critical failures in Meta's ad moderation systems at a time when the company is already facing a federal trial and multiple lawsuits. The incidents underscore the need for stronger platform accountability and more robust safeguards against AI-generated harm. Whether Meta can regain trust will depend on whether it takes meaningful corrective action — and whether regulators force it to.