An entire generation of children and adolescents is growing up conversing, learning, creating, and seeking advice from artificial intelligence systems that respond almost instantaneously. Far from a distant futuristic projection, this shift is happening rapidly in homes and classrooms worldwide.

According to UNICEF data published in 2026, children are adopting AI technologies more than three times faster than adults. Representative surveys across ten countries indicate that at least 20 million minors aged 12 to 17 have used AI, with approximately 13 million utilizing it for schoolwork. However, a more pressing statistic reveals that about two million minors turn to AI for advice on personal matters and concerns before speaking with parents, teachers, or friends.

The Dual Nature of Educational AI

Experts emphasize that artificial intelligence should not be viewed solely as a threat. When implemented thoughtfully, AI can serve as a powerful educational instrument. It offers patient explanations, translates content, adapts exercises to individual student levels, and provides immediate feedback.

Research from the Brookings Institution notes that well-designed tools can enrich learning within a solid pedagogical strategy. The dividing line emerges when AI transitions from a tool that assists learning into a substitute for the cognitive process itself. Asking an algorithm to help clarify a difficult concept is fundamentally different from instructing it to complete an assignment entirely.

Insights from PISA 2025 and the Risk of Lost Effort

Data from the PISA 2025 assessment—which surveyed over 760,000 15-year-old students across 91 countries and economies—offers an initial look at how students integrate digital tools and AI into their studies. While the findings point primarily to associations rather than direct causation, students who rely more frequently on AI for academic tasks tend to show less favorable educational outcomes.

Andreas Schleicher, Director of Education and Skills at the OECD, compares intellectual growth to physical conditioning: just as watching sports does not improve physical fitness, true learning requires the "productive cognitive struggle" of connecting ideas, making mistakes, and working through problems. When technology bypasses that struggle, it risks fostering a generation capable of gathering answers quickly while diminishing their ability to construct them independently.

A Developmental Perspective Across Childhood

The impact of artificial intelligence varies dramatically depending on a child's age and cognitive development. Research published in *Pediatrics* by investigators affiliated with Children’s Hospital of Philadelphia highlights that the implications of generative AI differ significantly between early childhood, middle childhood, and adolescence.

To safeguard healthy development, families and educators can apply three guiding principles:

  • Think before asking: Students should attempt to reason and formulate answers independently before consulting AI to check or expand their work.
  • Consulting does not mean believing: Minors must learn to cross-reference AI-generated outputs with reliable sources and expert knowledge, recognizing that fluency does not equate to truth.
  • Adults belong in the room: Pedagogy, upbringing, and emotional guidance cannot be outsourced to algorithms; teachers and parents must remain actively engaged in understanding what tools children use and why.

Deciding the Future of a Generation

Ultimately, the goal is not to reject artificial intelligence, nor is it to hand over the responsibilities of education and emotional support to machines. The central challenge is ensuring that technology adapts to children, rather than forcing childhood to adapt to technology.

Whether young people utilize AI to think critically or rely on it to replace human thought will not be determined by an algorithm. That outcome rests entirely in human hands.