Taking place annually on the second Tuesday of February, this year’s 10th February marks the 2026 Safer Internet Day. The Safer Internet Day, celebrated in over 160 countries, serves as a global prompt to promote positive internet use, raise awareness of online safety for children and young people, and reflect on how digital environments can be made safer, more inclusive, and more empowering by listening to children’s experiences and perspectives. At the EU level, the initiative is embedded within the broader Better Internet for Kids (BIK) framework, which provides the strategic and policy guidance for creating a better internet for Europe’s children and youth. Safer Internet Centres are responsible for the operational implementation of Safer Internet Day at the national level, adapting messages and activities to local contexts.
Each year, Safer Internet Day highlights emerging online issues and evolving risks, from cyberbullying and digital identity to social networking and platform design. The theme of this year’s Internet Safer Day edition is “Smart teach, safe choices – Exploring the safe and responsible use of AI”, which invites reflections on the growing role of generative artificial intelligence (AI) in shaping young people’s online experiences. In particular, it brings light to how rapidly evolving technologies such as generative AI will affect young people’s online experiences, in terms of benefits and risks, with a look to how AI-powered tools will change the ways in which young people create, learn, and communicate, how can users identify content that has been generated or altered by AI, and how scammers and hackers would modify their malicious behaviours accordingly.
The internet plays a big role in the lives of children. In 2024, the share of young people using the internet daily ranged between 93% and 100% across EU countries.[1] Children and young people access internet and online platforms for socialising, education and learning opportunity reasons, playing and entertainment, expressing creativity, online shopping, and for information and civic participation. Minors are also increasingly accessing emerging AI-enabled tools and platforms such as ChatGPT and Snapchat’s MyAI, for similar reasons like information searches, learning and education purposes, social connection, creativity and self-expression, and accessibility/automatic translations. By 2025, around 75% of young Europeans had tried generative chatbots.[2] National surveys show AI adoption among minors is already very high (around 88–92% in Slovenia[3] and 92,5% in Italy[4] claim to have used it at least once).
While online platforms and AI-enabled tools provide a variety of services and opportunities, they also present consistent adverse consequences deriving from the way that services are used (e.g., exposure to harmful content, allowing for abusive interactions, etc.) and from platforms’ design choices (e.g., recommender systems, addictive features, weak privacy protections).
Online, minors are exposed to significant risks that can undermine their well-being, safety and social and cognitive development. These risks relate to:
- Exposure to harmful content, violence and abuse: including gender-based violent content – leading to the normalisation of violence against women and girls, cyberbullying, misinformation, pro-anorexia – due to exposure to unrealistic body images – or self-harm material, and harmful AI-generated content. The use of the Internet to perpetrate or facilitate the sexual abuse and exploitation of children is a widespread issue. One in five children in Western Europe experiences unwanted or pressured sexual interactions online before the age of 18.[5] Girls are generally more affected by online gender-based violence. Further, digital technology has facilitated the rise of a new phenomenon, i.e. the “cyber-dating abuse”, whereby young women are subjected to constant surveillance and control by their partners. Also, the porn industry has been radically reshaped by new technologies and by social media, enabling the rapid production and distribution of both commercial and amateur pornography across pornographic platforms and social media. This shift has contributed to the spread of increasingly extreme, degrading and violent content, generally perpetrating patriarchal dynamics of society, therefore mainstreaming toxic masculinity in sexual culture and exacerbating gender-based violence as a consequence.
AI-generated materials further enhance the risks posed by online platforms. AI‑generated child sexual abuse material (CSAM) experienced an increase of over 1,325% in a single year (2025),[6] which is then subsequently disseminated through online platforms. Systemic risks affecting children safety and well-being are also posed by generative AI chatbots, with which children increasingly often create emotional attachment that may disrupt their relational development.[7] Lastly, deepfakes and synthetic media are increasingly used to disseminate misinformation, gender-based violence, and harmful health advice.[8] - Privacy and data exploitation: default privacy protection settings rarely protect minors: geolocation, microphone, and camera features are frequently active by default, allowing malicious actors to track or monitor children.[9] Likewise, authentication mechanisms and age-assurance systems (e.g. “I am 18” self-verification model) may not be the best option to ensure protection of children rights.[10] Despite platforms’ terms and conditions restricting use to those over 18, enforcement is often weak, leaving minors exposed. Algorithmic amplification, recommender systems, and inadequate content moderation further exacerbate online risks encountered by minors.[11]
- Addictive and manipulative design practices: Social media platforms employ design features, such as autoplay, infinite scroll, popularity metrics, and push notifications, that encourage prolonged engagement and compulsive use.[12] Also, a few online platforms expressed concerns regarding manipulation taking place of their services. In particular, they highlighted adults posing as minors, and creating fake profiles, with the intention of grooming them or manipulating children to lead them to share private information/materials, or recruit them for organised crime or terrorism. Children and adolescents are particularly vulnerable to manipulation and polarisation, as their cognitive and emotional skills for distinguishing reliable from unreliable information are still developing.[13] This is even more concerning considering that children access digital media at increasingly younger ages.[14]
The below figure highlights some of the problematic situations minors have encountered online, according to a survey conducted across a few EU countries.
Figure 1: Situations faced by minors online
Source: Growing up Online (Euroconsumers Survey)[15]
Addressing the risks that minors face online, and preventing the rise of new ones especially caused by the massive use of AI-enabled tools, is of the utmost importance considering the severe consequences. Mental health has declined significantly in minors, characterised by feelings of isolation – mood swings and depression, suicidal thoughts – sedentary lifestyles and lower physical well-being, attention problems, eye strain, headache, sleep deprivation, musculoskeletal issues, as well as language and cognitive deficits.[16] Also, content shared on social media, doxxing – the practice of revealing and publicising the records of an individual, which were previously private or difficult to obtain – revenge porn, can cause long-term reputational harm. Filter bubbles can lead to polarisation of thoughts, and narrow minors’ worldview. Remarkably, children with lower digital skills are even more exposed to such risks. They can be more easily victims of abuse and/or manipulated – being less able to recognise fake content and generally the risks they run in the online environment. They may also be less aware of reporting mechanisms in case of abuse – which may not already be child-friendly, widely known or efficient.
Many of the questions raised on the occasion of this year’s Safer Internet Day edition closely align with the obligations introduced by the Digital Services Act (DSA), which requires online platforms to identify, assess, and mitigate systemic risks, including those affecting children. At Open Evidence, we conducted research on systemic risks linked to Very Large Online Platforms (VLOPs) and Very Large Online Search Engines (VLOSEs) design, content moderation, and algorithmic systems. The study, finalised in July 2025, informed the first report of the Board of Digital Services and the European Commission, providing an overview of the most prominent and recurrent systemic risks stemming from the design, functioning and use of the VLOPs and VLOSEs, the key risk factors identified, and the existing and proposed risk mitigation measures.
On 14 July 2025, the European Commission adopted guidelines to support compliance with Article 28 of the DSA, which sets out specific obligations for online platforms to ensure a high level of privacy, safety, and protection for minors. The guidelines provide practical direction on measures such as age-appropriate design, risk mitigation, and the assessment of how platform features may affect children’s well-being. Open Evidence is currently conducting a three-year longitudinal study aimed at designing a monitoring framework to support the enforcement of Article 28 DSA and related provisions on the protection of minors. The study aims to track benefits, harms, and risks encountered by children in the online environment over time.
Further supporting the implementation of the DSA, the EU age verification blueprint initiative (still under development), aims at establishing an EU-harmonised approach to age verification to ensure that, across the whole EU, individuals prove to be 18+ to legally access adult-restricted online content, such as pornography, gambling, purchasing alcohol, and others. Importantly, the solution would also be easily adapted to prove other age ranges, to facilitate checking age gates required to access social media and other platforms.
Discussion among the EU institutions is also ongoing on the proposed EU Regulation on combating child sexual abuse online (so-called “child sexual abuse regulation”), which in November 2025 found a common negotiating position on a draft regulation across EU Member States’ representatives in the Council. The Draft Regulation is designed to create a long-term legal framework obliging digital service providers to take proportionate action to prevent the dissemination of CSAM and the solicitation of children through their services, through risk assessment, periodic reporting and cooperation with authorities as well as the establishment of a new EU Centre on Child Sexual Abuse.
In addition to EU-level policy developments, some EU Member State governments are exploring or advancing age-based restrictions on social media access. For example, the French National Assembly adopted a bill that would ban social media use for children under 15. The legislation has passed the lower house and is moving toward the Senate.[17] Similarly, Spain has officially announced that it will prohibit access to social media for minors under 16, requiring digital platforms to implement “effective age-verification systems” and take stronger action to prevent abuse[18] and Slovenia is preparing legislation to ban social media access for children under 15.[19] Discussions of similar bans are also ongoing in Denmark and Germany. Outside Europe, Australia has enacted one of the world’s first comprehensive social media restrictions for minors: the Online Safety Amendment (Social Media Minimum Age) Act 2024 establishes a minimum age of 16 for holding an account on major social media platforms.[20] Under the law, age-restrictedsocial media platforms[21] must take “reasonable steps” to prevent Australians under 16 from creating or maintaining accounts or face significant fines. In the US, federal legislation has been proposed in 2025, the Kids Off Social Media Act (S. 278), that would prohibit users under 13 from accessing social media platforms and also limit use of personalised recommendation systems for users under 17.[22] The bill passed out of the Senate Commerce Committee and is pending further consideration in Congress.
Author: Giorgia Spolverato
References:
[1] Young people – digital world – Statistics Explained – Eurostat
[2] Youth Internet Monitor (2025). Young People’s Online Use in Austria & Europe. Saferinternet.at / Better Internet for Kids initiative. Available at: Youth Internet Monitor 2025.
[3] Almost all teenagers in Slovenia already use AI – most use it for school, some for entertainment
[4] https://tg24.sky.it/cronaca/2025/11/14/adolescenti-intelligenza-artificiale-italia-dati
[6] Ibid.
[7] European Commission (2025). First report of the European Board for Digital Services in cooperation with the Commission pursuant to Article 35(2) DSA on the most prominent and recurrent systemic risks as well as mitigation measures. Available at: Press statement of the European Board for Digital Services following its 16th meeting | Shaping Europe’s digital future
[8] Vosoughi, S., Roy, D., Aral, S. (2018). The spread of true and false news online. Science, 359(6380), 1146-1151. Available at: The spread of true and false news online | Science; Karagianni, M. Doh (2024). A feminist legal analysis of non-consensual sexualized deepfakes: contextualizing its impact as AI-generated image-based violence under EU law. Porn Studies, 1–18. Available at: https://doi.org/10.1080/23268743.2024.2408277; Maras, M., Logie, K. (2024). Countering the complex, multifaceted nature of nude and sexually explicit deepfakes: an Augean task?. Crime Sci 13, 31. Available at: https://doi.org/10.1186/s40163-024-00226-6.
[9] Ramokapane, K. M., Mazeli, A. C., Rashid, A. (2023). Skip, Skip, Skip, Accept!!!: A Study on the Usability of Smartphone Manufacturer Provided Default Features and User Privacy. arXiv preprint arXiv:2308.14593. Available at: [2308.14593] Skip, Skip, Skip, Accept!!!: A Study on the Usability of Smartphone Manufacturer Provided Default Features and User Privacy.
[10] 5Rights Foundation (2018). Disrupted Childhood. The cost of persuasive design. Available at: Disrupted childhood: the cost of persuasive design – University of Hertfordshire (Research Profiles); 5Rights Foundation (2024). A high level of privacy, safety & security for minors. Available at: A High Level of Privacy, Safety & Security for Minors – 5rights; BEUC (2025). Better Safe than Sorry. BEUC position paper on how to keep children safe online in the EU. Available at: Better Safe Than Sorry.
[11] Ramokapane, K. M., Mazeli, A. C., Rashid, A. (2023). Skip, Skip, Skip, Accept!!!: A Study on the Usability of Smartphone Manufacturer Provided Default Features and User Privacy. arXiv preprint arXiv:2308.14593. Available at: [2308.14593] Skip, Skip, Skip, Accept!!!: A Study on the Usability of Smartphone Manufacturer Provided Default Features and User Privacy.
[12] Amnesty International (2023). Driven into Darkness: How TikTok’s ‘For You’ Feed Encourages Self-Harm and Suicidal Ideation. Available at: Driven into Darkness: How TikTok’s ‘For You’ Feed Encourages Self-Harm and Suicidal Ideation – Amnesty International; EDRi (2024). Tech platforms must drop addictive features that harm young people. Available at: Tech platforms must drop addictive features that harm young people – European Digital Rights (EDRi); BEUC (2025). Better Safe than Sorry. BEUC position paper on how to keep children safe online in the EU. Available at: Better Safe Than Sorry.
[13] Tervo-Clemmens, B., Calabro, F. J., Parr, A. C. et al. (2023) A canonical trajectory of executive function maturation from adolescence to adulthood. Nat Commun 14, 6922. Available at: https://doi.org/10.1038/s41467-023-42540-8; JRC (2025). The JRC explains: Why are children and adolescents vulnerable to social media? Available at: https://joint-research-centre.ec.europa.eu/jrc-explains/why-are-children-and-adolescents-vulnerable-social-media_en; Haidt, J. (2024). The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press.
[14] Smahel, D., Machackova, H., Mascheroni, G., Dedkova, L., Staksrud, E., Ólafsson, K., Livingstone, S., and Hasebrink, U. (2020). EU Kids Online 2020: Survey results from 19 countries. EU Kids Online. Doi: 10.21953/lse.47fdeqj01ofo
[15] Euroconsumers. (2025). Growing up online: Building a digital future for minors, by minors (Report). Euroconsumers. Available at: https://www.euroconsumers.org/wp-content/uploads/2025/09/Growing-up-online-Building-a-digital-future-for-minors-by-minors-1-.pdf
[16] Bozzola, E., Spina, G., Agostiniani, R., Barni, S., Russo, R., Scarpato, E., Di Mauro, A., Di Stefano, A. V., Caruso, C. Corsello, G., et al. (2022). The Use of Social Media in Children and Adolescents: Scoping Review on the Potential Risks. Int. J. Environ. Res. Public Health 2022, 19, 9960. https://doi.org/10.3390/ijerph19169960; O’Neill, B. (2023). Research for CULT Committee – The influence of social media on the development of children and young people, European Parliament, Policy Department for Structural and Cohesion Policies, Brussels. Available at: The influence of social media on the development of children and young people; Boniel-Nissim, M., Marino, C., Galeotti, T., Blinka, L., Ozoliņa, K. et al. (2024). A focus on adolescent social media use and gaming in Europe, central Asia and Canada: Health Behaviour in School-aged Children international report from the 2021/2022 survey. World Health Organization. Regional Office for Europe. https://iris.who.int/handle/10665/378982; Haidt, J. (2024). The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness. Penguin Press; Capraro, V., Globig, L., Rausch, Z., Rathje, S., Wormley, A., Olson, J., Ross, R., …, & Van Bavel, J. J. (2025). A Consensus Statement on Potential Negative Impacts of Smartphone and Social Media Use on Adolescent Mental Health. Seton Hall Law School Legal Studies Research Forthcoming, HEC Paris Research Paper No. MKG-2025-1567, Available at: https://ssrn.com/abstract=5256747; JRC (2025). The JRC explains: Why are children and adolescents vulnerable to social media? Available at: https://joint-research-centre.ec.europa.eu/jrc-explains/why-are-children-and-adolescents-vulnerable-social-media_en.
[17] Frances National Assembly approves ban on social media for under-15s
[18] Spain plans to prohibit social media use for under-16s; El Gobierno español prohibirá el acceso a redes sociales a menores de 16 años | ORAIN
[19] Frances National Assembly approves ban on social media for under-15s
[20] Online Safety Amendment (Social Media Minimum Age) Act 2024 – Federal Register of Legislation; https://www.esafety.gov.au/about-us/industry-regulation/social-media-age-restrictions
[21] The list of social media age-restricted platforms is regularly updated and available at: Which social media platforms are age-restricted? | eSafety Commissioner
[22] https://www.congress.gov/bill/119th-congress/senate-bill/278/text