Most conversations about AI in education focus on high school and university students. That makes sense. Those are the age groups writing essays with ChatGPT, navigating AI-generated misinformation, and preparing for a workforce shaped by automation. But there’s a growing body of research suggesting we should be thinking about AI literacy much earlier. I’m talking about children aged 3 to 8.
Su, Ng, and Chu (2023) conducted a scoping review of empirical studies on AI literacy in early childhood education. They examined 16 studies published between 2016 and 2022, looking at how young children are taught about AI, how learning is assessed, and what outcomes have been reported. The results are encouraging, and they challenge the assumption that AI literacy is too complex for young learners.
I’ll say upfront: I advocate for embracing AI widely, aggressively, and unapologetically across education. The early years? I am not really sure and I am still learning more about the potential of AI with young learners. And this study definitely helps!
I know young children are already living with AI. Voice assistants, recommendation algorithms, smart toys, all part of their daily world. So the question many ask is whether we help them understand what they’re interacting with, or leave them to form misconceptions on their own.
How Young Children Learn About AI
Most successful interventions in the review rely on playful, activity-based, experiential formats. Children work with robots, conversational agents, and simplified machine learning tools. Concrete experiences anchor abstract concepts in ways that feel natural and age-appropriate.
PopBots, for example, introduces rule-based systems, supervised learning, and generative AI through games like Rock-Paper-Scissors and classification activities. Conversational agents like Zhorai let children “teach” a system and observe how training data affects outputs. Su et al. describe the pedagogical grounding as an approach that “exploits both collaborative interaction and access to information-rich resources” (p. 6). Constructivist to the core.
And it works, as the authors state, “kindergarten children demonstrated understanding of rule-based systems as well as supervised machine learning” (p. 9). Generative AI , however, remains harder for this age group, which isn’t surprising. But foundational concepts like patterns, data, and learning from examples are within reach when the teaching is playful and well-designed.
What excites me is the broader skill development. Some studies reported growth in inquiry skills, collaboration, and theory of mind. Children who engaged with AI-interfaced toys didn’t just learn about AI. They learned to explore, question, and express their understanding through play. Su et al. describe AI perspectives as “attitudes and dispositions adopted while solving problems” (p. 11). At this level, AI literacy is as much about curiosity and reflection as it is about technical content.
The Teacher Readiness Problem
Now for the part that worries me. Teacher knowledge and confidence come up repeatedly as major barriers. According to Su et al., “teacher AI knowledge, in particular, was discovered to be one of the challenges encountered by educators who do not have technical background” (p. 10).
This isn’t unique to early childhood. The pattern shows up at every level. Celik (2023) found in his Intelligent-TPACK study that technical knowledge alone doesn’t predict effective AI integration. Pedagogical judgment and ethical awareness are stronger predictors. Choi, Jang, and Kim (2023) found that ease of use was the strongest driver of teacher AI acceptance, and that trust requires transparency and sustained experience. Bilbao-Eraña and Arroyo-Sagasta (2025) showed that even brief AI literacy training can shift awareness and attitudes, but building genuine confidence takes time and depth.
Early childhood educators face the same dynamic. They want to introduce AI concepts but don’t feel prepared. Curriculum frameworks, clear teaching guidelines, institutional support, all missing or underdeveloped. Su et al. flag every one of these as structural barriers.
My take: we can’t wait for perfect readiness. Give early childhood educators playful tools, simple lesson structures, and professional development that builds confidence gradually. The children are ready. The pedagogy exists. The bottleneck is support.
What’s Missing From the Research
Su et al. acknowledge significant gaps. There’s “no standard questionnaire, survey or test for assessing young children’s AI knowledge/skills” (p. 10). Assessment instruments are still being developed, and most studies rely on pre/post knowledge tests or observational methods. The field is young. Exploratory.
I’d add another gap: affective outcomes. Almost all studies focus on cognitive gains, what children know or can do after an intervention. Very few measure confidence, long-term motivation, or ethical awareness. These are important, especially if we’re trying to build AI literacy as a lifelong disposition, something children carry with them as AI becomes more embedded in their world.
And then there’s equity. Children from lower socioeconomic backgrounds sometimes show lower familiarity with coding tools, which affects AI-related experiences. Roe, Furze, and Perkins (2025) argued in their Critical AI Literacy paper that any framework for AI in education must account for access and power. Early childhood AI literacy is no different. If we only develop these programs in well-resourced schools, we widen the gap before it even begins.
Reference
- Bilbao-Eraña, A., & Arroyo-Sagasta, A. (2025). Fostering AI literacy in pre-service teachers: Impact of a training intervention on awareness, attitude and trust in AI. Frontiers in Education, 10, 1668078. https://doi.org/10.3389/feduc.2025.1668078
- Celik, I. (2023). Towards Intelligent-TPACK: An empirical study on teachers’ professional knowledge to ethically integrate artificial intelligence (AI)-based tools into education. Computers in Human Behavior, 138, 107468. https://doi.org/10.1016/j.chb.2022.107468
- Choi, S., Jang, Y., & Kim, H. (2023). Influence of pedagogical beliefs and perceived trust on teachers’ acceptance of educational artificial intelligence tools. International Journal of Human–Computer Interaction, 39(4), 910–922. https://doi.org/10.1080/10447318.2022.2049145
- Roe, J., Furze, L., & Perkins, M. (2025). Digital plastic: A metaphorical framework for Critical AI Literacy in the multiliteracies era. Pedagogies: An International Journal. Advance online publication. https://doi.org/10.1080/1554480X.2025.2557491
- Su, J., Ng, D. T. K., & Chu, S. K. W. (2023). Artificial intelligence (AI) literacy in early childhood education: The challenges and opportunities. Computers and Education: Artificial Intelligence, 4, 100124. https://doi.org/10.1016/j.caeai.2023.100124