The quality of early childhood education at Mardhatilah Kindergarten remains low due to limited use of interactive technology and insufficient teacher understanding of innovative digital media. This community service project aims to enhance teachers’ instructional quality by integrating deep learning technology and digital literacy to create more adaptive and engaging learning environments. The study employed a Participatory Action Research (PAR) approach, consisting of diagnosis, participatory planning, action, observation, and reflection. Teachers actively participated in designing AI-based learning media, including visual–auditory object recognition supported by deep learning. The findings indicate a significant improvement in teachers’ digital competencies, particularly in operating AI-assisted learning tools. This advancement contributed to increased student engagement and improved comprehension through interactive AI-generated visualizations. Furthermore, the integration of deep learning with digital literacy enabled teachers to transform conventional teaching methods into more effective and relevant practices aligned with contemporary educational demands. In conclusion, the integration of deep learning and digital literacy through PAR effectively empowers early childhood educators to develop innovative, adaptive, and high-quality learning environments. The synergy between artificial intelligence and creative pedagogy is essential for sustainably improving early childhood education standards.
Copyrights © 2026