Child safety

Children are increasingly exposed to serious online harms, including abuse, eating disorders and dangerous self-harm content. CCDH researches how platforms fail to protect young users, and what must change to build a safer internet for kids and teens.

Take action

Research

Our research for this topic covers various issues.

View all research

Up Next: Anorexia Algorithm

How YouTube still recommends eating disorder videos to young girls

In February, a landmark trial in Los Angeles found YouTube’s parent company, Google, liable over claims that its addictive platform harmed a young woman, known as KGM. She started using YouTube at six years old and later developed an addiction to social media that contributed to body dysmorphia and mental health issues. YouTube’s algorithms have the power to shape what people are watching and can seriously harm vulnerable users. CCDH tested the platform’s protection in the US, UK, and EU, and found that it still recommends harmful eating disorders videos to accounts registered as 13-year-old girls. Our investigation also found that YouTube can reduce these harms. Harmful eating disorder recommendations fell from an average of 1 in 3 videos in 2024 to 1 in 9 in 2026. The improvements followed new safety obligations under the UK’s Online Safety Act (OSA) and the European Union’s Digital Services Act (DSA), showing that social media platforms can address harms when lawmakers hold them accountable. Progress matters. But one harmful recommendation to a child is one too many. YouTube must stop recommending harmful eating disorder content altogether.

View report Published

Killer Apps 

How mainstream AI chatbots assist users planning violent attacks

8 in 10 AI chatbots were regularly willing to assist users in planning violent attacks including school shootings, religious bombings, and high-profile assassinations. DeepSeek went as far as wishing the would-be attacker a “Happy (and safe) shooting!”. These are the findings of our new report based on research conducted in collaboration with CNN’s investigative unit. These digital prompts don’t stay online. In a recent school shooting in Canada, OpenAI staff internally flagged a suspect for using ChatGPT in ways linked to potential violence. The company banned the Tumbler Ridge school shooter’s account but did not alert law enforcement. Months later, that user allegedly killed eight people and injured at least 25. The guardrails exist. Most companies are choosing not to use them, putting public safety and national security at risk.

View report Published