Content area
This dissertation presents a historical analysis of artificial intelligence integration in Western-educational paradigms from the 1960s to 2022, examining how the persistent gap between AI's promised transformational impact and its consistently modest educational outcomes reveals fundamental tensions between computational and pedagogical logics. Using qualitative historical methodology, this study analyzed primary and secondary sources across four distinct eras of AI development in North American and European contexts. The analysis reveals three critical findings. First, AI's educational impact has remained remarkably consistent across technological generations—positive but modest, with effect sizes typically between 0.2 and 0.8—suggesting that fundamental constraints transcend specific technological limitations. Second, implementation factors prove more determinative of success than technological sophistication. Third, each era reproduces similar patterns of promise-disappointment cycles, equity challenges, and tensions between efficiency and educational values, indicating systemic rather than technical challenges. The study identified a persistent "complexity gap" between what AI can computationally model and what education requires humanistically. Successful implementations have consistently been those that enhanced rather than replaced human expertise, suggesting that the future of AI in education lies not in automation but in augmentation. These findings suggest that realizing AI's educational potential requires understanding AI as one tool among many for addressing specific educational challenges within existing institutional and cultural contexts.