Background and Rationale

Rapid advances in AI technologies, such as generative AI and increased adoption of AI systems across the globe, have resulted in proliferation of human rights risks. Children as early and enthusiastic adopters rs of these systems, are particularly vulnerable, Children and young people are increasingly relying on AI products and services, not only as go-to sources for information, learning and creativity, but also for advice, support and sometimes relationships. The myriad of human rights risks pertaining to use of AI products and services include but are not limited to: discrimination and bias, including datasets contaminated with harmful and illegal content; the environmental impacts of AI systems, which particularly impact children; AI-generated child sexual abuse material (CSAM) and non-consensual intimate images (NCII); and the effects of AI on children when used in armed conflict and cyber operations.

While AI is increasingly becoming a part of children’s lives, despite the world’s focus on regulation and governance guidelines on AI, opportunities for children to effectively voice their concerns and contribute to the development of appropriate safeguards are scarce.  This significant gap has prompted regulators and policymakers to take urgent actions, regarding child safety measures in the use of digital and AI products and services by children. However, a piecemeal approach to rights-based AI governance across geographies has resulted in a fragmented AI governance landscape. There is an urgent need to shift the AI governance approaches from reactive to preventive, to ensure that AI innovation advances by adopting a child-rights based approach where child safeguards are not an afterthought but the starting point.

As AI continues to evolve, its development, deployment, and use carry important human rights implications, especially for children who are the lead adopters of AI products and systems. It is imperative to leverage an ecosystem approach with roles and responsibilities for a range of stakeholder groups from government, businesses, civil society, communities, parents/caregivers and educators, to ensure the adoption of a child rights-first approach to  AI product lifecycle including design, development and deployment.

AI solutions that prioritize human rights protections are not only safer but also more trustworthy, fostering consumer confidence and strengthening the long-term viability of AI-driven advancements. Organizations are missing out on critical opportunities to do so due to the absence of robust HRDD during the product design and development stage. By embedding fairness, transparency, and accountability into AI systems, businesses can create technologies that serve society and are safe for all, while also sustaining their competitive edge.

A proactive and systematic approach ensures that AI products are both groundbreaking and responsible, balancing technological advancement with fundamental human rights protections.

RBHR Form Question: How does your session connects to the theme of Building Resilience, Advancing Rights?

Advancing Rights: Under the UNGPs, the corporate responsibility to respect human rights is operationalized through human rights due diligence (HRDD) which is an ongoing and dynamic process. Despite growing awareness of human rights considerations among AI developers, practical guidance on how to undertake HRDD during the development phase of AI products, with a special focus on safeguarding children’s rights, is limited. There is an urgent need to integrate HRDD throughout the product design and development process.

"Under the UNGPs, the corporate responsibility to respect human rights is operationalized through human rights due diligence (HRDD) which is an ongoing and dynamic process through which companies, including AI tech companies : 1. identify and assess actual and potential human rights risks 2. take action to prevent and mitigate adverse impacts 3. track the effectiveness of responses, 4. communicate how impacts are addressed. 

in the context of digital tech, HRDD must be continuous, forward-looking and embedded across the entire tech Lifecyle. When it comes to AI specifically, HRDD should 1. begins at the design phase ("human rights by design") this is the core of the session 2. considers impacts across the full lifecycle of the technology and 3. remains ongoing and responsive to new and emerging risks.

Building Resilience: At the same time, there is a critical need to strengthen the capacity of various stakeholders (Investors, civil society organizations) to engage with AI development as critical partners. To provide meaningful oversight and feedback, these groups must understand not only the potential human rights risks of the AI systems at hand, but also the technical processes, decision points, and governance structures that shape how these products are designed and deployed.

By equipping these actors with practical knowledge of embedding human rights and child rights safeguards in the AI product development lifecycle, we can foster more inclusive dialogue, enable richer and decision-useful stakeholder input, and ensure that risk identification and mitigation strategies are informed by external experts and rightsholder representatives. This will not only enhance the quality and legitimacy of HRDD but also ensure that AI systems are developed in ways that mitigate risks and take into account the needs and priorities of specific use groups and communities, particularly for their leading adopters and the most vulnerable.

Convening Partners and Complementary Workstreams

This session integrates converging and mutually reinforcing workstreams:

B-Tech 2.0: AI Product Design Workstream: OHCHR B-Tech Project and Business for Social Responsibility (BSR) joint project on AI Product Design Workstream in cooperation with AI industry leaders. The workstream aims to develop practical, foundational guidance on how human rights due diligence can be integrated into AI product development processes, particularly for downstream AI products that are customized or fine-tuned for specific contexts and uses.

UNICEF AI for children: UNICEF is leading global efforts for AI to be designed and governed in the best interests of every child. Through research, policy guidance and advocacy, UNICEF is working to urgently close the participation and evidence gaps, and to place children’s rights at the heart of the AI age.

Guiding Questions for Discussion

The session is structured around the following thematic questions, which will frame the panel discussion and stakeholder interventions:

• Data governance and privacy, including the lawful, necessary and proportionate use of personal data (especially children’s data in AI training and deployment);

• Risk identification in contexts of uncertainty, including the application of precautionary approaches where evidence of harm is still emerging (e.g. generative AI systems)

• Potential long-term and systemic impacts, including on mental health, development, access to information, and discrimination;

• Transparency and accountability, including enabling independent research and communicating how risks are being addressed.