Parents who rely on a platform’s safety settings to decide whether their child is ready for social media may be trusting tools that do not hold up under testing, according to a new report from the Cybersafety Research Center, an independent group that reviews child safety claims made by technology companies.
According to the Cybersafety Research Center’s analysis, researchers tested 86 child safety features across Instagram, TikTok, Snapchat, and YouTube, comparing what each company claims its tools do against how those tools performed in practice. The center found that 51 of those features, or about 60 percent, failed, while only 35 features, just over 40 percent, both worked as described and were something a young user could realistically find and use.
The Cybersafety Research Center said it judged each feature on two measures: whether it functioned as the platform claimed, and whether a child could actually locate and use it without adult help. Based on this method, the center reported failure rates of 73 percent on Snapchat, 66 percent on Instagram, 55 percent on YouTube, and 50 percent on TikTok, with Instagram having the largest number of features tested at 29.
Some of the examples the report describes point to gaps parents may not expect. Per the report, a TikTok test account registered to a minor searched for material related to disordered eating, and the app’s own search function suggested related terms, including phrases about hiding food and self-harm. The report also states that on Instagram, a similar search for eating-disorder content prompted the app to offer alternate search terms with deliberate misspellings that could get around the platform’s restrictions, and that misspellings worked to bypass restrictions on Snapchat as well. Separately, the Cybersafety Research Center raised concerns about “take a break” reminders on Instagram, YouTube, and TikTok, noting that each included an option to dismiss the prompt and keep scrolling.
The platforms disputed these findings when contacted by CNN, which first reported on the study. A TikTok spokesperson told CNN that its teen accounts include more than 50 preset safety features turned on automatically, along with additional parental controls through its Family Pairing tool, and said an internal review confirmed the features work as intended. A Snapchat spokesperson told CNN that the company cares about the safety and wellbeing of its users and has worked for years to build safeguards, and said many of the report’s findings relied on deliberate attempts to bypass protections rather than typical use. According to CNN’s reporting, Meta and YouTube said their time-limit and break features work as intended, and noted that limits set through parental controls cannot be dismissed the way in-app reminders can.
The report arrives against a backdrop of mounting legal exposure for the platforms. In March 2026, a Los Angeles jury found Meta and YouTube liable in a case alleging their platforms were designed to be addictive and had contributed to a young user’s mental health struggles, according to court reporting from CNN and NPR, ordering the companies to pay several million dollars in damages; both companies said at the time that they would appeal. Thousands of similar lawsuits remain pending against the four companies, which continue to dispute the underlying claims.
None of this means safety settings are worthless, and the Cybersafety Research Center’s report does not argue that parents should abandon them. What it does suggest is that these tools should not be treated as a finished job. Reviewing privacy settings, messaging permissions, search filters, and screen-time controls directly inside each app, rather than assuming the advertised defaults are working, remains a reasonable step for families. So does the more basic work no report can replace: talking with children about what they encounter online, who contacts them, and how they feel about what they see, on a basis that does not depend on any single feature functioning correctly.
