Social media algorithms expose Irish children to harmful content despite age restrictions
A CyberSafeKids report found that Irish children aged 8 to 12 continue to access age-restricted social media platforms and encounter harmful content including pornography, scams, and false information, despite age-assurance measures introduced in July 2025. The report documented rapid growth in children's use of generative AI and noted that disparities exist between children in DEIS and non-DEIS schools in their exposure to online harms.
What happened
A report by CyberSafeKids released in September 2026 found that many Irish children aged 8 to 12 continue to access 13+ social media platforms despite age-assurance measures that came into effect in July 2025.
The report documented increased exposure of children to scams and false information online. CyberSafeKids stated that social media platforms were pushing pornography towards underage children, with access to sexually explicit content significantly increasing over the past year despite European rules requiring age verification for children under 18 to access such material.
Use of generative AI among young people has increased substantially, with four times the number of 12- to 15-year-olds and double the number of 8- to 12-year-olds using the technology compared to the previous year.
Reported impact
Children reported continued negative experiences online including access to scams, fake information, and pornography despite age restrictions.
The report found stark disparities, with children in DEIS schools experiencing higher rates of underage access to social media, AI chatbots, and contact with strangers online compared to non-DEIS schools.
Index analysis
CyberSafeKids CEO Alex Cooney stated that social media platforms know the real age of children using their services but are not adequately protecting them, and that platforms could implement age verification as a definitive measure.
The charity criticised amendments to EU AI legislation as containing 'get-out clauses' that weaken protections, such as allowing AI chatbots to avoid disclosing their nature if it is 'obvious to a reasonable person', creating ambiguity and potential for legal challenges.
Governance controls
Age-assurance measures came into effect in July 2025 under legal obligations, but the report found these have failed to reduce underage access to age-restricted platforms.
Proposed solutions such as parental controls have been suggested but researchers questioned their effectiveness, stating that the industry had already relied on parental controls for social media without solving underlying harms.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.