Should we allow unregulated AI use by minors?
In recent years, artificial intelligence (AI) systems like ChatGPT and Gemini have become part of daily life for adults and children. Once unimaginable, these technologies have now become tools for students to seek help with homework, entertainment and even emotional support. However, minors are using these systems without understanding how they work. Allowing unregulated AI exposure at such a vulnerable stage of development is not only irresponsible but also dangerous.
Earlier this month, the Interscholastic Ethics Society (IES), a student-led organization representing over 150 students from 11 international schools in Korea, submitted a national petition urging the government to implement regulations on the use of generative AI by children under the age of 12. Their petition, posted on the National Assembly’s public platform, calls for three measures: Limiting AI access for children under 12 (with exceptions for supervised educational use); requiring safety features such as age verification, suicide-related dialogue blocking and usage-time limits; and providing AI-safety education for schools, teachers, and parents. They believed these are reasonable safeguards that acknowledge the realities of children and adolescents using digital technology today.
IES’s argument is based on a concerning trend. More than half of Korean adolescents currently use generative AI tools, and research shows that a significant number trust AI-generated information without question. Children, who at a stage when imagination is strong, are especially prone to accepting misinformation, emotionally manipulative content or seemingly authoritative AI responses because they are delivered in fluent and confident language.
Overseas, there have been cases of minors suffering serious emotional harm after prolonged interactions with AI chatbots, with some incidents even leading to legal action. Furthermore, research from such top institutions as MIT and Harvard has raised concerns that excessive reliance on AI can lead to memory loss, impaired analytical thinking and mental health issues. While these findings do not imply that AI is inherently harmful, they demonstrate that we have introduced disruptive technologies into children’s lives without establishing meaningful safeguards.
Other regions already recognize the issue. Both the U.S. and the European Union restrict AI use by children under 13 and require parental consent for minors. Despite its technological advancements and high digital literacy, Korea lacks similar safeguards for generative AI. This creates a regulatory vacuum, allowing young children to freely access tools designed for adults that can shape their beliefs, emotions and behaviors in unpredictable ways.
IES leaders stressed in an official statement that their goal is not to impede technological advancement. Connor Seong, IES senior secretary, stated, “We are not against AI. We are against unregulated exposure that could harm children who do not yet understand the risks.”
This distinction is crucial. AI will play a vital role in future education, innovation and problem-solving. However, technological advancement does not justify abandoning fundamental responsibilities for child safety.
IES President Joseph Kang emphasized this point even more directly, stating, “The safety of young people should not depend on algorithms.” In an era where companies design AI systems to maximize engagement, not child development, leaving safety to private platforms is not enough. Regulation is not a barrier but rather a support system that enables constructive use of AI and prevents avoidable harm.
The student petition is particularly noteworthy because it is led by young people. This reflects a generation that understands the benefits of AI while also recognizing its risks —perhaps more clearly than many adults. The fact that these students have taken it upon themselves to advocate for national change underscores the urgency of this issue. When young people demand safeguards, policymakers, educators and parents have a responsibility to listen.
As AI permeates every aspect of learning and social interaction, we cannot leave children to navigate this environment alone. Children need tools designed with their vulnerabilities in mind, including systems that incorporate basic safeguards such as age-appropriate content, restrictions on harmful conversations and the guidance of trained adults.
Unregulated use of AI by minors is a societal risk, and this risk will only intensify as AI becomes more sophisticated. The question is not whether children will use AI, but whether we can ensure that they use it safely. The petition submitted by these students is a timely and responsible call to action. Korea and all societies grappling with the rise of generative AI should heed this call.
Pae Hee-kyoung (Global Youth Institute) is an educator (admin@gyi.co.kr).