Analysis of the use of artificial intelligence in mathematics education in Libya

Abstract

Artificial intelligence is entering mathematics education, but evidence about its use in Libya remains uneven. This study examines adoption, teaching applications, and learning claims across publicly available Libyan research. A targeted bilingual review identified eight relevant reports published between 2024 and September 2026. These included mathematics surveys, wider university surveys, documentary studies, and one reported teaching experiment. Published tables were extracted and checked through descriptive calculations. International experiments provided comparison evidence, without serving as estimates for Libya. The selected mathematics surveys covered Gharyan, Zawia, and Al-Marj. The Al-Marj study included 337 teachers, with primary teachers forming 74.2% of its sample. Reported classroom-use scores were lower than several workload and curriculum-barrier scores. A broader Zawia survey reported substantial training needs, alongside less frequent assessment use. However, most published percentages could not be reconstructed using its stated total sample alone. Another Zawia study contained a scale summary inconsistent with its displayed item means under ordinary averaging. A reported linear-equation experiment showed a nine-point difference in score changes between groups. Its reporting limitations prevented a firm causal interpretation. Overall, the evidence supports local engagement with AI, but cannot establish national adoption or sustained learning benefits. The paper proposes teacher-reviewed hints, independent mathematics assessment, and transparent reporting as priorities. Future Libyan studies should measure actual use, delayed learning, access differences, and teaching costs separately.