40% of child protection resources on the internet do not work, says study

Social media giants have for years touted their growing suite of tools and security protections as proof that they prioritize the well-being of young users. But more than half of these protections don't work as advertised, new research finds.

Researchers at the Cybersafety Research Center tested 86 youth safety features on TikTok, Instagram, Snapchat and YouTube, and analyzed whether they worked as described and whether children would be able to find and use them realistically. Only 35 of these features - just over 40% - successfully met both criteria.

YouTube, Instagram and Snap have widely disputed the report's conclusions, arguing that their features work as intended or that the tests did not represent typical use of the platforms by children and teens. TikTok did not respond to CNN's request for comment on the report. YouTube and Instagram's parent company Meta were also found responsible this year for intentionally addicting and harming young people.

All four companies face thousands of lawsuits with similar allegations, which they have disputed.

"If you are a parent, you may know that we found systemic problems in the design and implementation of many of these features," says the report, published on Monday. For the Cybersafety Research Center, a joint initiative of New York University and Northeastern University, the findings show that the harm of social media to children "is not hypothetical, and that, when it occurs, the consequences can be irreversible."

The findings call into question social media companies' claims that major investments in new tools and features have made their platforms safe for young people. And they come amid renewed pressure for more federal regulation of social media, with executives from some major companies expected to be summoned to testify on Capitol Hill again this summer.

The tests
The researchers created two sets of test accounts - those with birth dates associated with minors to test child safety features, and adult accounts aged 25 and over to test restrictions on interactions with children. They tested whether a young person using the platforms would normally find the features, whether a teenager trying to get around them would be successful, and whether an adult user would be able to bypass restrictions on messaging to minors.

Features were classified as failures when they: were buried in settings and were difficult to find, did not work to prevent damage as advertised, or both. Nine features were also labeled as "missing" because researchers were unable to activate them even when they tried. For example, comments between teenage Instagram accounts that included bullying language - such as profanity and "nobody (sic) likes you" - did not trigger the platform's announced warning for the user to rethink their comment.

Instagram stated that the "pause to rethink" warning was not designed to appear when the posting user and the commenting user follow each other.

"To be effective, security features need to be on by default or easy to turn on, be resistant to typical teen use, and must be proven to protect against harm," the report states.

The report did not analyze parental control features, reviewing only tools that would apply directly to children and teenagers or that could be activated by them. The platforms claim that parental controls also contribute to youth safety and make it difficult for children and teens to circumvent time and other resource restrictions.

The number of features available for testing varied by platform, with Instagram having the most, with 29. Feature failure rates by platform were: Snapchat 73%, Instagram 66%, YouTube 55%, and TikTok 50%.

The Resources
All four platforms claim to block children from searching for dangerous content and instead direct them to supportive resources. But the researchers found that, in practice, this fell short of expectations.

After a test TikTok account registered in the name of a minor searched for material about eating disorders and self-harm, the app's search function began suggesting terms such as "how to pretend to eat your food" and "razor blade skin," according to the report. When a child test account started typing "eating disorder" into the Instagram search bar, the app automatically offered alternative search terms with deliberate misspellings that could bypass content restrictions. Misspellings also worked to bypass restrictions on Snapchat.

A TikTok spokesperson said the app's teen account settings "come with more than 50 preset features and security settings automatically enabled, with additional options for parents through our easy-to-use Family Pairing tool. Our internal review confirms that these features are working as intended, and we appreciate the opportunity to help the authors of this report better understand how our app works."

Researchers called YouTube's search content restrictions a success. But they said YouTube's efforts to redirect young users to supportive resources failed because a child user could click away from the restricted screen and continue viewing the content they had searched for. YouTube says that if a teen dismisses this asset targeting screen, they will still only see content that is rated safe for their age group.

"We've spent more than a decade building industry-leading parental controls, which is why 84% of parents who have used YouTube's supervised account tools said they agree that these tools give them confidence that their children are accessing a safer, more controlled digital environment," a YouTube spokesperson said in a statement about the report. "We will continue to strengthen these protections and innovate to protect families who use YouTube."

Researchers also raised questions about Snapchat and Instagram's efforts to block unknown adults from contacting young users via messaging. On Snapchat, researchers using an adult test account said they were able to find and send messages to a child account without any restrictions. The child's account "received the friend request and, upon accepting it, was able to view the message history that the adult had sent, without any warning," the report states.

On Instagram, researchers found that adults a child didn't follow were unable to start messaging conversations with that child, saying the app's restriction works as described. But they argued that the feature is compromised because a child can send messages to an adult they don't follow and the adult can then respond - without restrictions or warnings.

The report points to a 2021 press release from Meta, which states that Instagram restricts people over the age of 19 from sending private messages to teens who don't follow them. "We found that this is not strictly true; an adult can message a child without restriction after contact has been initiated, even if the child does not follow back," the report states.

Meta told CNN that a teenager messaging an unfollowing adult indicates that the young person wants to connect. The researchers say they alerted Instagram and Snapchat to concerns about messages from children and other "critical vulnerabilities" before releasing the report.

The researchers say they alerted Instagram and Snapchat to concerns about messaging from children and other "critical vulnerabilities" before releasing the report.

A Snapchat spokesperson told CNN that the company cares "deeply about the security, privacy and well-being of all Snapchatters, and our teams have worked for years to build safeguards, release security tutorials and partner with experts...we are continually evaluating and strengthening our protections." The spokesperson added that many of the report's conclusions were based on "intentional actions to circumvent protections that are not representative of the typical user experience."

Many of the platforms also offer time limits or "take a break" reminders to prevent children from scrolling endlessly. But on Instagram, YouTube and TikTok, the prompt telling users to take a break includes an option to "snooze" the reminder and return to the feed. Meta and YouTube stated that the feature works as intended and added that time limits set through parental controls cannot be waived.

"This report is fundamentally flawed and demonstrates a basic misunderstanding of how our tools work," a Meta spokesperson said in a statement. "Authors include vague claims that our features are faulty, but in the vast majority of cases they either misrepresent those features or fail to provide examples or evidence. The reality is that with Teen Accounts, teens are viewing less sensitive content, having fewer unwanted contacts, and spending less time on Instagram at night."

Meta argued that some of the report's conclusions about "missing" features may have occurred simply because researchers did not find issues that would have triggered the tools. For example, the researchers said they saw no "sensitive content" warning screens on potentially problematic posts; Meta stated that no sensitive content may have appeared in its feeds.

Some appeals were successful. For example, if users under the age of 13 try to sign up for an account on TikTok, they are automatically directed to a "TikTok for Younger Users" experience, which is view-only and removes potentially harmful app features like search and messaging.

This means that children do not have the opportunity to try to register again with a false, older date of birth on that device.

Minors' accounts are automatically set to private on Instagram, which researchers say "could encourage users to keep accounts private without requiring the user to make an informed decision."

These advances prove that it is possible to develop effective safety resources for young people, researchers said. Companies should create safer online platforms to reduce overall risk, they wrote, "rather than filter out a dangerous experience."

Why is social media under fire for being "addictive"?

Source: CNN