Instagram accounts that follow preteen influencers are fed an algorithm of sexual adult content and risqué footage of children, according to a report.
The Wall Street Journal published an investigation Monday into Instagram’s algorithm for its Reels feature, which automatically collects a continual series of quick videos that users may be interested in, similar to TikTok. The WSJ investigation focused on what algorithm is produced for users that “follow only young gymnasts, cheerleaders and other teen and preteen influencers active on the platform.” The results were heavy doses of adult content mixed with suggestive footage of children.
The investigation was conducted based on an observation that young influencers on Instagram often have a sizable following from adult men, and many of these adult men focus their feeds on sexual content.
The WSJ set up the test Instagram accounts on newly purchased devices. The algorithms apparently became increasingly sexual as the accounts expanded their following from the preteen influencers to the followers of those influencers.
One specific succession of videos went as follows: an adult content creator uncrossing her legs to reveal her underwear, a sprinter at a track meet running over a boy, an ad for Disney, an adult content creator in lingerie with a furry tail and blood dripping from her mouth, a child in a bathing suit recording herself in a mirror, another adult content creator, and a girl shaking her buttocks to music in a car.
The Canadian Centre for Child Protection ran similar but separate tests of its own — and its findings concurred with the WSJ’s. The group concluded that Instagram’s Reels often portrayed clothed children who appear in the National Center for Missing and Exploited Children’s digital database of images. Some videos were confirmed to be child abuse sexual material.
Meta, the parent company of Instagram that also owns Facebook, contended that the WSJ’s investigation focused on manufactured accounts, so the results did not reflect most users’ experiences. The company did not comment on why the algorithms became so sexual.
Current and former Meta employees reportedly told the WSJ in anonymous interviews that algorithms promoting child sexualization content were internally known to be a problem. They apparently said attempts to counter this content from spreading to relevant users conflict with policies that prevent major changes to recommendation algorithms. The WSJ reviewed company documents on the matter to confirm the complaints.
Samantha Stetson, VP of client council and industry trade relations at Meta, said the feature underwent rigorous testing before its full-scale rollout.
“We tested Reels for nearly a year before releasing it widely, with a robust set of safety controls and measures,” she said.
But a Meta safety analysis reportedly advised that Reels would link videos of children and inappropriate content, according to the WSJ. The company’s former head of youth policy, Vaishnavi J, summarized the recommendation given by Meta’s safety staff to the WSJ as: “Either we ramp up our content detection capabilities, or we don’t recommend any minor content.”
The WSJ’s newly published investigation was a follow-up to its report in June that found Instagram had a significant community of users interested in pedophilic content who used the platform to communicate. The platform’s algorithm would connect pedophiles and recommend them to content sellers, as The Dallas Express reported at the time. Meta apparently set up a task force after the story was published and began removing tens of thousands of accounts monthly.
Lianna McDonald, executive director for the Canadian Centre, said online communities focused on child sexual abuse have found ways to thrive through social media algorithms.
“Time and time again, we’ve seen recommendation algorithms drive users to discover and then spiral inside of these online child exploitation communities,” McDonald told the WSJ.
Another concern detailed in the investigation focused on how major companies had ads featured between Instagram Reels of adult sexual content. Most companies request in their advertising agreements with social media companies to stay separate from adult content.
Several of Instagram’s advertisers gave statements to the WSJ expressing concern over the investigation. Match, an online dating website company, stopped advertising through Reels in response.
“We have no desire to pay Meta to market our brands to predators or place our ads anywhere near this content,” Match spokeswoman Justine Sacco told the WSJ.
Some fringe advertisers that showed up on Reels for the WSJ’s accounts included live-streaming adult websites, massage parlors with “happy endings,” and artificial-intelligence chatbots meant for cybersex.
Meta told companies who complained about advertisements for their brands being juxtaposed with adult content that it was investigating the issue and would enlist brand-safety auditing services to examine how frequently this occurred, per the WSJ.