Meta’s ad review systems are failing to catch the rampant child sex abuse material (CSAM) being featured, promoted, and sold on their platforms. And the world’s most vulnerable children are suffering the consequences.
Last weekend, BBC India released an alarming report alleging that Instagram’s automated ad review systems were monetizing and pushing CSAM on its platform. According to the BBC, their team created a test account on the Meta-owned app after noticing that Instagram was pushing highly suggestive sexual content even when users are not actively searching for it.
After clicking on some of these posts, the algorithm reportedly started serving the BBC’s test account with paid ads for pornography, including content that appears to feature adults with minors. Once users clicked on these links, they were directed to Telegram channels, where more content was offered and sold. BBC encountered at least 30 unique ads promoting CSAM. Even more alarming, when they tried flagging some of the posts through the platform’s in-app reporting tool, they received a message 24 hours later, saying that the content did not violate community guidelines and would remain online.
This latest revelation adds to the growing concerns over how children and teens are being exploited and trafficked through social media platforms, with the majority of these alleged cases taking place on the platforms owned by Meta. Since 2023, The Guardian has published a series of investigative reports on how Facebook and Instagram have allegedly been used to groom, recruit, buy, and sell children. In one report, former contract workers, who were tasked with moderating the platform, claimed that although they flagged and escalated possible child trafficking activity, these complaints “often went nowhere,” and that harmful content was rarely taken down.
The accounts are highly disturbing. Offenders described how easy it is for them to identify which children and teenagers are most vulnerable to grooming based on their online activity. They would then establish contact with their target through Meta’s direct messaging features. Detailed stories from survivors reveal similar patterns wherein perpetrators build trust by grooming them with constant compliments and attention. They are then expertly coerced and manipulated into being trafficked once that trust has been earned.
While many of these reported cases involved other countries, the Philippines cannot afford to treat this as a distant issue. A 2022 Unicef study found that about two in 10 internet-using Filipino children aged 12 to 17 had experienced online sexual abuse and exploitation in 2021. That translates into roughly 2 million children. Even more concerning is that only 0-4 percent of the victims reported it to authorities, and only 0-3 percent knew how to report it to the authorities or proper helplines. This suggests that the problem might be happening at an even larger scale.
Part of the urgent solution is, of course, to equip and empower children with the skills to identify risks, protect themselves, and seek help when they encounter online harm. The same study found that 44 percent of children do not know where to get help if they or a friend are subjected to sexual abuse or harassment online, 50 percent of them added people as friends on social media even if they had never met face-to-face, and 13 percent of these children eventually met their new online friends in person.
Through the years, Meta has announced various initiatives to improve online safety, including the use of artificial intelligence to detect harmful content. Admittedly, the scale of the problem makes it difficult to create a perfect solution. But the findings of BBC and The Guardian’s reports suggest that even the most basic safety mechanisms and reporting systems are failing. In contrast, abusers and organized crime networks seem to be having an easy time taking full advantage of existing platform design features and paid advertising tools to amplify and distribute harmful content.
This issue comes at a time when stricter social media rules, including a possible ban for underage users, are being discussed by Philippine public officials. Apart from sexual content, the wake of the Tacloban shooting has exposed how children may be increasingly vulnerable to online grooming for violence and extremism.
Yet in many of these discussions, much of the accountability is still placed on users and their parents. A ban on young people’s access to social media may help shift the culture around online behavior, but it will not fully address the deeper issue. These social media platforms are hosting harmful content, are allegedly being used to promote, recommend, and monetize it, and have not invested enough resources to address it. The solution cannot simply be to restrict children’s access and tell their parents to be more careful online. It must include sustained pressure on social media companies to make their platforms safer and more child-friendly by design.
If these platforms are sophisticated enough to predict what will enable their users to keep watching, sharing, and buying, then they should also be compelled to invest resources in ensuring their systems will stop children from being exploited, advertised, and sold.
—————-
eleanor@shetalksasia.com