Preparing Learners and Teachers for an AI-Driven Future: Emerging Trends, Pedagogical Challenges, and Critical Perspectives in Pre-University AI Education: A Systematic Literature Review

Synthesis: Caruana, Gilar-Corbí, and Palomar (2026) present a PRISMA-guided systematic literature review of 42 studies (2009–2026) on AI integration in pre-university (K-12) education, motivated by Sustainable Development Goal 4 and the principles of Education for Sustainable Development. Thematic synthesis identifies four major research areas: teacher education for AI integration (n=16), curriculum development and pedagogical foundations (n=13), AI literacy and ethical competencies (n=15), and the pedagogical and ethical implications of generative AI (n=9). The evidence consistently indicates that effective AI integration hinges on teacher preparedness, structured interdisciplinary curricula, and critical, ethical approaches that foster responsible digital citizenship — with significant gaps in longitudinal evidence, classroom-based empirical research, and standardized AI-literacy assessment instruments.

Key Findings

1. Teacher education is the linchpin. The most consensus-heavy finding is that effective and responsible AI integration depends on teachers' pedagogical and technological preparedness. AI-focused professional development is positively associated with instructional quality (cognitive activation, classroom management, personalized learning), and frameworks such as Intelligent-Technological Pedagogical Content Knowledge (TPACK) and AI-TPACK predict teachers' readiness, with AI literacy, algorithmic ethics, and digital competence as core ingredients.
2. Structured, interdisciplinary curricula are needed. Research supports frameworks like AI4K12 and the Five Big Ideas in AI. AI education is shifting from narrow technical content toward competency-based models that incorporate critical thinking, digital citizenship, Creativity, problem solving, and ethical reflection — with emerging competencies like prompt engineering gaining recognition.
3. AI literacy is a cross-cutting competency. AI literacy combines technical understanding, critical thinking, algorithmic ethics, and responsible digital citizenship. The review calls for embedding it as a cross-curricular competency, though standardized assessment instruments remain lacking.
4. Generative AI offers opportunities and raises tensions. GenAI tools (ChatGPT, Gemini, Copilot) support retrieval, content generation, personalized tutoring, and instructional planning, but raise concerns about cognitive offloading, content reliability, pedagogical oversight, and technological dependency.
5. Equity and governance are cross-cutting. Disparities in access, connectivity, and training risk reinforcing educational inequalities. The review calls for governance models, institutional policies, and regulatory frameworks ensuring transparency, accountability, human oversight, data protection, and academic integrity.

Pre-university AI education as a distinct focus

This review consolidates the growing evidence that AI education in primary and secondary settings is distinct from higher education. It centers on teacher preparation, curriculum design, and the age-appropriate development of AI literacy and ethical competencies — aligning with the knowledge base's K-12 coverage. Notably, the review is framed by Sustainable Development Goal 4 (SDG 4) and Education for Sustainable Development (ESD), situating equitable, resilient, and sustainable educational systems as the goal of responsible AI integration. (Note: this framing draws on SDG 4 but the review's substance is K-12 AI education rather than AI's environmental footprint.)

Research gaps

• Scarcity of longitudinal studies on the sustained effects of AI on learning outcomes, cognitive autonomy, and socio-emotional development.
• Predominance of exploratory/conceptual work over authentic classroom-based empirical research.
• Geographical concentration in Europe and North America; limited representation from developing regions and from primary (vs. secondary) education.
• No standardized instruments for assessing AI literacy, critical thinking, or ethical GenAI competencies.

What this means for practice

• Instructors. Invest in AI-focused professional development that pairs AI literacy and algorithmic ethics with motivational and institutional factors (Self-Efficacy, perceived usefulness, organizational support), since teacher preparedness is the review's most consensus-heavy condition for effective integration.
• Instructors. Embed AI literacy as a cross-curricular competency rather than a standalone technical subject, covering technical understanding, critical thinking, algorithmic ethics, and responsible digital citizenship.
• Instructors. Rethink assessment beyond final products toward complex reasoning and reflective engagement with AI, given the documented risks of cognitive offloading, content unreliability, and technological dependency.
• Instructors. Prepare to teach structured, interdisciplinary curricula such as AI4K12 and the Five Big Ideas in AI, including competency-based models that add critical thinking, digital citizenship, creativity, problem solving, and ethical reflection alongside prompt engineering.
• Instructors. Work within — and press for — governance that ensures transparency, accountability, human oversight, data protection, and academic integrity, because access, connectivity, and training disparities otherwise reinforce existing inequalities.

Limitations

• The search was restricted to English-language publications, which may have excluded relevant research from other linguistic and educational contexts.
• The corpus spans 42 studies of mixed type (empirical, conceptual, and review articles), making direct comparison difficult and allowing overlap in the underlying evidence, so the findings are a qualitative thematic synthesis rather than independent evidence.
• No formal methodological quality or risk-of-bias appraisal was conducted: all included studies contributed equally regardless of design, and the review was not prospectively registered.
• Google Scholar served as a complementary source whose selected results depended on the platform's ranking algorithms, introducing retrieval bias; evidence is additionally concentrated in Europe and North America and thin on primary (versus secondary) education, and GenAI's rapid evolution may date parts of it quickly.

Citation

Caruana, I., Gilar-Corbí, R., & Palomar, M. (2026). Preparing Learners and Teachers for an AI-Driven Future: Emerging Trends, Pedagogical Challenges, and Critical Perspectives in Pre-University AI Education: A Systematic Literature Review. Sustainability, 18(17), 8827.