Meta details its ad-review and enforcement steps after reports of Instagram ads allegedly violating child-exploitation policies

Meta says it took swift steps after media reports alleged that Instagram ads in India violated the company’s policies against child exploitation. In a statement posted to its newsroom on July 7, Meta said it had “already identified and disabled several of the violating ads and the accounts behind them” before the issue was brought to its attention.

The company said its follow-up review led to additional enforcement, including removing more ads, disabling accounts, and blocking URLs linked to content that it says violated its rules. Meta also rejected the suggestion that it would knowingly and deliberately target such ads to people based on “an inappropriate interest in children,” saying the company uses technology to spot accounts showing potentially suspicious activity related to children and then removes them.

Meta offered a quantitative measure of that enforcement approach, saying it automatically removed “over 4 million” accounts last year that it characterized as potentially related to child exploitation. It said criminals can try to evade detection, and that while its review process may not catch every violation, it is continuously upgrading both automated and human review steps to stay ahead of repeat offenders.

The statement describes how Meta’s advertising enforcement is designed to act before ads run and also to re-check them over time. According to Meta, the ad review system automatically checks ads for policy violations before they are approved for delivery, and people can also report ads if they believe they violate Meta’s advertising or community rules. Meta said ads “remain subject to review and re-review at all times,” and that ads can be rejected or restricted at any point in the process.

Meta said the review covers all components of an ad, including images, video, and text, and it also evaluates the advertiser’s broader presence. If a violation is found during the review process, Meta said it can reject the ad and restrict the advertiser’s Business Account or its assets, such as Pages and user accounts. If a business account or its assets are restricted, Meta said they cannot be used to advertise across Meta’s technologies.

The company also outlined its stance on policy scope, saying it maintains “detailed and robust policies” covering child nudity, abuse, and exploitation. It tied those policies to its broader advertising standards, saying ads must comply with its Community Standards on Child Sexual Exploitation, Abuse, and Nudity, and must not include content that “sexually exploits or endangers children.”

Meta linked its enforcement approach to a wider program of proactive detection and stepped-up use of artificial intelligence. It said it had previously shared positive results from changes that focused proactive enforcement toward illegal and the “most severe” content, including child exploitation. It also said it has been experimenting with more advanced AI systems for content enforcement across languages spoken by 98% of people online, expanding beyond coverage that previously covered around 80 languages.

As an illustration of how such alleged activity can reach users through advertising funnels, reporting earlier this week by BBC Eye found Instagram running paid adverts in India that it described as promoting child sexual abuse material and linking users to channels on Telegram. The BBC said the ads used terms including “rape” and “child video,” and that Instagram’s moderation technology approves posts before they run, while Meta later told the outlet it had disabled ads and accounts in response to the investigation. Meta did not provide details in its July 7 statement about the specific ads at issue in the BBC reporting.

Meta’s statement also points to industry and government coordination as part of its strategy. It said that in 2019 it played a role in establishing a relationship between India and the U.S. National Center for Missing and Exploited Children (NCMEC), enabling local law enforcement to receive reports and take action. The company framed that as part of strengthening reporting frameworks that support others in combating exploitation.

Meta did not disclose additional operational details in its post, such as the number of ads removed in this specific case, the geographic scope of the enforcement actions, or the precise mechanisms used to detect the advertiser accounts flagged as potentially suspicious. It also did not provide a clear timeline matching the enforcement steps to any particular news report, beyond reiterating that some action occurred before the company said it was notified by outside parties. For investors and regulators watching Meta’s safety posture, the company’s latest disclosure is more about process than metrics tied to the India incident itself.