A new study by Cybersafety Research Center has found that more than half of the child safety features on major social media platforms do not work as promised, raising fresh questions about how well children are protected online.

Alongside this, an Australian study found that social media companies are failing at the very first step of checking users' ages, making Australia's teen social media ban much less effective. Australia introduced a world-first law in December that requires platforms like Instagram, Snapchat and YouTube to stop children under 16 from having accounts, according to Reuters.

Under the law, platforms must take "reasonable steps" to check users' ages, and the government has advised companies to use multiple age-check methods instead of relying on just one system. However, several studies suggest that many children under 16 are still able to access social media platforms, leading Australia to double the maximum fines last month and warn companies they could face court action.

The companies disputed many of the findings, saying the features work properly or that researchers did not test them in a way that reflects normal teen behaviour. The report also noted that Meta and YouTube were found liable this year of intentionally addicting and harming young people, while all four companies continue to face thousands of similar lawsuits, which they deny.

"If you are a parent, you should know that we have found systemic issues with the design and implementation of many of these features", Cybersafety Research Center report states. It also says social media harms to children "are not hypothetical, and that when they do occur the consequences can be irreversible."

The Cybersafety Research Center study, a joint initiative of New York University and Northeastern University, tested 86 child safety features across TikTok, Instagram, Snapchat and YouTube. Researchers checked whether the safety features worked as described and whether children could easily find and use them. The study found that only 35 out of 86 safety features—just over 40%—met both standards, as per CNN. That means more than half of the tested child safety features failed.

One example involved Instagram, where bullying comments such as curse words and "no body (sic) likes you" did not trigger the platform's warning asking users to rethink the comment, according to CNN. Instagram responded that the "pause to rethink" prompt is not designed to appear when both users follow each other.

The report states, "Safety features, in order to be effective, need to be on by default or easy to activate, be resilient to normal teenager use, and should demonstrably protect against harm."

On TikTok, a child account searching for eating disorder and self-harm content later received search suggestions including "how to pretend to eat your food" and "razor blade skin", according to the report.

On Instagram, typing "eating disorder" suggested alternate spellings that could help users bypass content restrictions. Researchers found that misspelled search terms also bypassed Snapchat's restrictions.

Also read: DeepSeek AI chip: China startup develops custom chip to reduce reliance on Nvidia and Huawei

Researchers said YouTube's search restrictions largely worked. However, they found children could dismiss the support screen and continue viewing the searched content. YouTube said users would still only see age-appropriate material.

A YouTube spokesperson said, "We've spent over a decade building industry-leading parental controls, which is why 84% of parents who have used YouTube supervised account tools said they agree that these tools give them confidence that their child is accessing a safer and more controlled digital environment", according to CNN. The spokesperson added, "We will continue to strengthen these protections and innovate to protect families who use YouTube."

Researchers also found that an adult Snapchat account could locate and message a child account without restrictions. According to the report, the child account "received the friend request and upon accepting it was able to view the history of messages that the adult had sent them with no warnings."

On Instagram, adults could not start conversations with children who did not follow them. However, if the child sent the first message, researchers found the adult could reply without warnings or restrictions. The report says, "The adult is able to send messages to a child unrestricted after contact has been initiated, even if the child does not follow them back." Meta told CNN that if a teen messages an adult first, it shows the teen wants to connect, so the feature is working as intended.

A Snapchat spokesperson said the company cares "deeply about the safety, privacy, and well-being of all Snapchatters, and our teams have worked for years to build safeguards, launch safety tutorials, and partner with experts … we are continually evaluating and strengthening our protections." The Snapchat spokesperson also said many findings came from "intentionally taking actions to bypass protections that are not representative of the typical user experience", according to CNN.

Researchers also found that Instagram, YouTube and TikTok allow users to snooze "take a break" reminders and continue scrolling. Meta and YouTube said those reminders work as intended, while stricter parental time limits cannot be dismissed. A Meta spokesperson said, "This report is fundamentally flawed and demonstrates a basic misunderstanding of how our tools work."

Instagram also automatically makes minor accounts private by default, which researchers said helps protect children without requiring them to change settings themselves. Researchers said these successful features show that effective child safety tools are possible if platforms design them properly.