5Rights Foundation

The first Global Dialogue on AI Governance must explicitly consider children's rights and their best interests, ensuring that AI systems uphold children's dignity, safety, and agency. AI systems are shaping children's experiences, choices, and futures in profound ways—governance mechanisms must recognize and respond to these realities. In line with the UN Secretary-General's call to mainstream children's rights throughout the work of the UN, and building upon global developments – such as on UNCRC General comment No. 25 which affirmed that children's rights apply fully in the digital world, and urgent calls for action throughout the UN system, including HRC Resolution 53/29 on New and Emerging Digital Technologies and Human Rights; UNGA Resolution 78/187 on the Rights of the child in the Digital Environment; the UN Secretary-General's High-level Advisory Body on AI's Governing AI for Humanity; and the Joint Statement on Artificial Intelligence and the Rights of the Child – the Global Dialogue should prioritize children's rights and best interests throughout their work. This, too, should be reflected in their establishment and functioning. The Dialogue should equip UN member states with the tools and frameworks necessary to fulfil their commitment in the Global Digital Compact to "[s]trengthen legal and policy frameworks to protect the rights of the child in the digital space." This requires establishing clear guardrails for AI design, deployment, and accountability; ensuring AI systems align with the UN Convention on the Rights of the Child (UNCRC) and its General comment No. 25 on children's rights in the digital environment. AI governance must not be reactive; it must anticipate risks and set proactive standards to ensure AI systems respect and uphold children's rights by design and default.

• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Protection and promotion of human rights
• Transparency, accountability, and human oversight

These areas are heightened for children, who are developing and acquiring the necessary knowledge, experiences, and resources to understand or challenge automated decision-making, algorithmic unfairness and the subtle, cumulative, or acute ways AI shapes their lives. This makes them uniquely vulnerable to harm when systems are not designed with their needs in mind. Considering AI-specific characteristics (e.g. opacity, complexity, dependency on data), AI systems that are not designed based on child-centered principles, trained on appropriate data sets, and tested to ensure neutral or positive outcomes for children can have significant impacts on children's rights safety, privacy, cognitive development, health and educational outcomes, social relationships, economic well-being and freedoms. Therefore, prioritizing safe, secure and trustworthy AI ensures that systems interacting with or impacting children are designed and tested to prevent harm from the outset. This must be grounded in the protection and promotion of human rights, including children's rights to privacy, safety, development, and participation. Transparency, accountability, and human oversight are essential to address imbalance of power, the inherent opacity and complexity of AI systems making the impact more severe, persistent, and difficult to detect. Taken together, these priorities support a Global Dialogue that must unite the global community to ensure AI is governed by a framework rooted in children's rights

Governance gaps in AI are already producing measurable harm for children, while allowing unsafe systems to scale globally without adequate scrutiny or accountability. Children are early adopters of new and emerging technologies, and AI is no exception: 1 in 3 users of digital products and services is a child, and children are twice as likely as adults to have used generative AI. From smart toys and personalized learning platforms to recommender systems, chatbots, and virtual assistants, AI is already shaping the environment in which children learn, play, and grow. Yet, in a context of a regulatory lag AI still lacks the most basic safeguards for children. Instead, it is pushed to market at breakneck speed, driving a dangerous race to the bottom where children and their safety are an acceptable collateral in the maximization of commercial profit. Children are already being exposed to these risks, from relentless privacy violations to recommender systems promoting self-harm and chatbots that nurture unhealthy emotional attachment and development. When companies operate without safeguards, the result is not innovation. It is harm at scale: manipulation, dependency, and threats to children's safety, privacy, and agency. We are already witnessing this: AI-related child safety risks have doubled between 2022 and 2025. (https://oecd.ai/en/incidents), underscoring both the urgency and the systemic nature of the problem. At the same time, there is a critical opportunity to develop concrete guardrails about how AI systems can operate. In line with the UN Secretary-General's call to mainstream children's rights across the work of the UN – and to fulfil the commitments under the GDC – it is imperative to ensure that children's rights and best interests are foundational principle in the functioning of the Global Dialogue and not an afterthought.

The Global Dialogue can play a decisive role in turning fragmented national efforts into coherent, rights-based international action. The UN Convention on the Rights of the Child is the most widely ratified human rights treaty in history. This demonstrates that there is broad and unwavering consensus that children must be protected. The Global Dialogue can have a crucial role in translating this global consensus into action by delivering governance frameworks that safeguard children in the age of AI. The Global Dialogue can drive be a catalyst for action and unite global efforts towards a convergence of standards to ensure that every child, no matter where they live, is protected by the same high standards. We cannot afford fragmentation. At a time when international cooperation on children's digital safety is more important than ever, recommitting to implementing the standards set out in UNCRC General comment No. 25 is an opportunity to build on global support and UN Member States' unanimous commitment to protecting children's rights in the digital world. By acting now and anchoring AI governance in the UN Convention on the Rights of the Child, we can build a future where AI serves children, instead of allowing tech companies to dictate the terms of our children's future.

Children's rights cannot be an afterthought in AI governance – they must be embedded from the outset. This requires the inclusion of experts in children's rights, neuroscientists, developmental psychologists, and child-centred AI design, as well as meaningful engagement with children themselves. Their voices and lived experiences must inform the Global Dialogue's work, ensuring AI systems are designed and deployed in ways that uphold their rights, dignity, and best interests.

UNCRC General comment No. 25 reflects the global consensus that all children hold the same rights and opportunities online as offline. It makes clear that children's rights are non-negotiable and must be the foundation of the digital world, not an afterthought. It also reemphasises tech companies' responsibility to ensure that technology impacting children is rights-respecting and age-appropriate by design and default. 5Rights' "Children & AI Design Code" provides a roadmap to embedding these standards into AI systems, allowing for innovation while ensuring that children's rights are at the heart of AI development. (https://5rightsfoundation.com/resource/children-ai-design-code/) The final report of the UN Secretary-General's High-level Advisory Body on AI - "Governing AI for Humanity" - similarly calls for strong protection and privacy for children, and mandates child impact assessments and age-appropriate design. (https://www.un.org/en/ai-advisory-body) Most recently, the "Joint Statement on Artificial Intelligence and the Rights of the Child" - endorsed by 13 UN agencies and international organisations - sets out concrete expectations for states and tech companies, building on UNCRC General comment No. 25 and echoing the standards set out in the 5Rights Children & AI Design Code. It notably makes clear that AI systems which are not rigorously tested before deployment to be safe or designed to respect children's rights must never interact with or impact children. (https://www.itu.int/hub/publication/d-str-cyb_joint-2025/) Finally, 5Rights' "Building a digital environment designed with children in mind" builds on international best practices that have emerged across every continent to set out a practical blueprint to address harmful design and business models, strengthen accountability, and ensure technology is safe and age-appropriate by design and default. (https://5rightsfoundation.com/resource/building-a-digital-environment-designed-with-children-in-mind-an-international-best-practices-blueprint/