USC Marshall School of Business Research Paper Sponsored by iORB, Forthcoming
Artificial intelligence (AI) innovation offers numerous benefits, yet its rapid development also presents significant risks, particularly for children.AI chatbots, powered by Large Language Models (LLMs), are becoming increasingly integrated into daily life, with platforms such as ChatGPT and Character.AI attracting hundreds of millions of users, including minors. While AI chatbots can provide mental health support and enhance communication skills, they also pose serious risks, including social isolation, exposure to child abuse, and even suicide.Given the global decline in youth mental health and the documented impact of technology (e.g., social media) on youth well-being, this issue demands urgent attention. Current AI governance frameworks often overlook the developmental needs and rights of children, failing to ensure that AI technologies foster human flourishing rather than cause harm. This paper argues that AI companies have both an opportunity and a responsibility to prioritize child well-being by designing chatbots that enhance, rather than replace, human relationships.To address these risks, we propose creating a Global Task Force on AI and Child Well-Being, led by G20 nations, to develop innovative standards for AI chatbot design and deployment. This initiative requires a collaborative, multi-sector approach to: 1. Developing an AI design paradigm that promotes children’s social and relational development2. Codifying this design paradigm into international technical standards and 3. Demonstrating its implementation through independent third-party testing. AI companies should show tangible benefits for youth before widespread deployment.By aligning AI development with principles of human flourishing, policymakers, and industry leaders can ensure that AI chatbots serve as tools for enrichment rather than sources of harm. This paper presents a roadmap for AI innovation that prioritizes child well-being.