AI literacy deficit threatens digital readiness in schools

Despite AI's rapid integration into classrooms, major gaps persist in how students and teachers understand, apply, and critically assess the technology, according to a new study published in Education Sciences.

The study, "Toward a Coherent AI Literacy Pathway in Technology Education: Bibliometric Synthesis and Cross-Sectional Assessment," combines a decade-long global literature mapping with a field assessment of AI literacy among secondary students and pre-service teachers. The findings provide an urgent call to align education systems with the competencies demanded by the age of intelligent machines.

AI literacy: The new cornerstone of digital education

AI literacy is now fundamental to preparing future citizens and professionals. The study brings together global AI literacy research from 2015 to 2025, drawing on the Web of Science database to identify leading themes, countries, and intellectual networks shaping the field. Using advanced bibliometric tools, they trace how the discourse has shifted from general discussions of artificial intelligence to concrete pedagogical frameworks for teaching AI concepts.

Their analysis shows that early studies focused largely on AI applications, but over the past five years, attention has turned toward human-centered competencies, including critical thinking, ethics, and human–AI collaboration. The rise of large language models and ethical AI frameworks has expanded the conversation to include bias, transparency, and academic integrity. These shifts suggest that AI literacy education has moved beyond coding and algorithms to encompass social, emotional, and moral understanding of intelligent technologies.

Key clusters identified in the mapping reveal three dominant research streams:

• Cognitive and technical literacy, focusing on how students grasp AI concepts and systems.
• Critical and ethical literacy, emphasizing awareness of bias, fairness, and accountability.
• Pedagogical integration, exploring how teachers and curricula adapt to AI-driven tools.

The study notes that these areas are becoming increasingly interconnected, signaling a global transition from isolated skill-based instruction to holistic AI competence development.

Testing AI literacy in real classrooms

To translate the bibliometric insights into real-world evidence, the authors conducted a cross-sectional assessment of AI literacy among 145 learners in Slovenia. Their sample included secondary technical students and pre-service technology and engineering teachers, tested during the 2024–2025 academic year using the standardized Hornberger et al. AI Literacy Test. The study provides one of the first data-backed comparisons of how different education levels and backgrounds influence AI literacy performance.

Results reveal a significant gap between students and future teachers. Pre-service teachers consistently scored higher across most AI literacy components, particularly in understanding intelligence and programmability, two skills crucial for effectively teaching AI principles. Statistical analysis confirmed a significant group effect (p = 0.02), indicating that teacher education programs are producing more advanced AI-literate individuals than technical secondary schools.

However, the findings also expose a concerning gender disparity. Males outperformed females overall (p = 0.01), highlighting a persistent imbalance in technology-related education that could widen if not addressed through targeted policy and curriculum reforms. Despite this, both groups showed similar proficiency in critical AI literacy, suggesting that awareness of ethical and societal issues surrounding AI is becoming more widespread across the educational spectrum.

According to the study, these differences highlight the need for structured AI literacy pathways, curricula that progress logically from foundational AI concepts to applied and ethical understanding. Such frameworks could help reduce disparities and ensure consistent competency development from secondary education through to teacher training.

Building the roadmap for future AI literacy

While AI is revolutionizing how people learn and work, education systems remain unevenly prepared. Their findings propose a four-pillar framework for coherent AI literacy in technology education:

• Foundational AI Knowledge: Teaching students how AI systems operate, including logic, data, and algorithms.
• Critical Appraisal: Developing the ability to question AI outputs, recognize bias, and understand ethical implications.
• Participatory Design: Empowering learners to co-create AI tools responsibly, not just consume them.
• Pedagogical Integration: Equipping teachers with strategies to embed AI thinking across all subjects, not just STEM.

The study also calls for policy-level interventions to embed these pillars in national education strategies. It notes that AI literacy must be approached as a lifelong learning process, supported by teacher professional development, cross-disciplinary collaboration, and curriculum modernization.

The authors recommend expanding future research to include larger and more diverse international samples, as well as longitudinal and mixed-method designs to track how interventions improve AI literacy over time. Furthermore, they urge policymakers to support AI literacy equity initiatives, particularly those addressing gender imbalances and the digital divide between well-resourced and underfunded schools.