Meta AI Ads Child Safety Crisis

San Francisco Moves Against Meta Over Controversial Meta AI Ads

AdTech

San Francisco orders Meta to halt Meta AI Ads linked to child abuse content after hundreds of harmful ads appeared across its social platforms.

San Francisco orders Meta to halt Meta AI Ads linked to AI-generated child abuse content.
Researchers identified 350+ harmful ads across Facebook, Instagram and Threads.
The ads reached more than 29,000 accounts and targeted users across several countries.
Meta says the ads were removed, but researchers report that new harmful ads continued appearing.
The case puts Meta Safety and Ad Moderation under pressure over how AI abuse content reaches users.

San Francisco’s City Attorney has ordered Meta to stop allowing Meta AI Ads linked to AI-generated child sexual abuse content and explain how the ads repeatedly slipped through its safety systems.

The cease-and-desist letter reported more than 350 harmful ads on Facebook, Instagram and Threads. The ads used images of minors, including some confirmed real people, and turned them into sexualized AI videos. Some also promoted AI apps that could digitally undress people or create nonconsensual intimate images.

How did so many harmful ads get past Meta’s safeguards?

The ads were identified by researchers at the Tech Transparency Project. They reached more than 29,000 accounts across European Union countries and also targeted users in the US, Australia and India.

More than 250 of the reported ads appeared after exposed 53 such ads in August. Some identical ads were also uploaded several times.

San Francisco City Attorney David Chiu accused Meta of failing to deal with a known AI Abuse problem while earning money from the advertising. His office says the repeated ads raise serious questions about Ad Moderation and Child Safety.

Meta has pushed back on the city’s authority. The company says there is no indication the ads appeared in San Francisco and argues that the matter is outside the city attorney’s jurisdiction.

But can an online platform really draw a city line around harmful content?

Meta says all the identified ads have now been removed for violating its policies. The company also said most had fewer than 200 impressions and that spending on the 300-plus ads was below $5,000. Meta maintains that it does not allow child exploitation, nudify apps or similar content.

The city wants more than takedowns. It has given Meta 28 days to explain how the ads bypassed Meta Safety systems, how reports are sent to the National Center for Missing and Exploited Children and how advertisers and repeat offenders are handled.

TTP says new harmful ads continued appearing even after its findings were reported.

If the same type of content keeps returning, are removals enough?

The dispute exposes a larger problem for Meta Ads. AI tools are making abusive content easier to create, while platforms must detect it before it spreads.

Business Honor examines Meta now faces pressure to show that its AI Content controls and safety systems can prevent these ads, not simply remove them after researchers or officials find them.