Instagram Teen Safety Features Found Flawed, Report Says

A report by child-safety advocacy groups and Northeastern University researchers has found that many of Instagram’s teen safety features are flawed or non-existent. Of 47 features tested, only eight were deemed completely effective.

Credit: META

Meta, Instagram’s parent organisation, disputed the findings, calling them erroneous and misleading. The report, titled "Teen Accounts, Broken Promises," compiled and analysed Instagram's publicly announced updates spanning over a decade.

Credit: META

Researchers noted that features designed to prevent young users from finding self-harm-related content were easily bypassed. Anti-bullying message filters also failed to activate, even when prompted with harassing phrases Meta had previously used to promote them.

Additionally, a feature meant to redirect teens from excessive consumption of self-harm content never triggered. However, some teen account safety features did work as advertised, including a "quiet mode" meant to temporarily disable notifications at night, and a feature requiring parental approval for changes to a child’s account settings.

Laura Edelson, a professor at Northeastern University who oversaw a review of the findings, stated that the results question Meta’s efforts "to protect teens from the worst parts of the platform." She added, "Using realistic testing scenarios, we can see that many of Instagram's safety tools simply are not working."

Two of the groups behind the report, Molly Rose Foundation in the United Kingdom and Parents for Safe Online Spaces in the U.S., were founded by parents. These parents allege their children died due to bullying and self-harm content on Meta’s platforms.

Meta spokesman Andy Stone said the report "repeatedly misrepresents our efforts to empower parents and protect teens, misstating how our safety tools work and how millions of parents and teens are using them today." Stone also called some of the report’s appraisals "dangerously misleading."

Stone contended that the company’s approach to teen account features and parental controls has evolved over time. He claimed that teens placed into these protections "saw less sensitive content, experienced less unwanted contact, and spent less time on Instagram at night."

Arturo Bejar, a former Meta safety executive and an Instagram consultant from late 2019 to 2021, provided tips to the advocacy groups and university researchers. Bejar, who left Meta in 2015 before returning as a consultant, stated he "experienced firsthand how good safety ideas got whittled down to ineffective features by management."

Bejar indicated that during his second tenure, Meta did not respond to data highlighting severe teen safety concerns on Instagram. Stone maintained that Meta responded to the concerns Bejar raised while employed by taking actions to make its products safer.

Reuters independently confirmed some of the report’s findings through its own tests and a review of internal Meta documents. One test showed that entering "skinny thighs" without a space, a term Meta had blocked, still surfaced anorexia-related content for a teen test account.

Internal Meta documents revealed that as the company promoted teen-safety features last year, it was aware of significant flaws. Safety employees had warned in the last year that Meta failed to maintain automated-detection systems for eating-disorder and self-harm content.

This failure meant Meta could not reliably avoid promoting content glorifying eating disorders and suicide to teens, nor divert users consuming large amounts of such material. Staffers also acknowledged that a system to block search terms used by potential child predators was not being updated promptly.

Stone stated that internal concerns regarding deficient search term restrictions have since been addressed by combining a newly automated system with human input.

The findings come amidst increased scrutiny on tech companies to safeguard young users. Last month, U.S. senators launched an investigation into Meta after Reuters reported on an internal policy allowing chatbots to "engage a child in conversations that are romantic or sensual."

This month, former Meta employees testified before a Senate Judiciary subcommittee, alleging the company suppressed research showing preteen users of its virtual reality products were exposed to child predators. Stone dismissed these allegations as "nonsense."

Meta announced Thursday an expansion of its teen accounts to Facebook users internationally. The organisation also said it would pursue new local partnerships with middle and high schools.

Credit: META

Instagram head Adam Mosseri commented, "We want parents to feel good about their teens using social media."

---

• A report by advocacy groups and Northeastern University researchers found Instagram's teen safety features are largely ineffective or flawed.

• Only eight of 47 tested features were fully effective, with many self-harm and anti-bullying tools failing to work as intended.

• Meta disputes the report's accuracy, with spokesman Andy Stone calling the findings "erroneous and misleading."

Source: REUTERS