New Perspectives in Science Education

The teaching of physics in upper secondary education frequently grapples with the need to render complex and abstract phenomena accessible and comprehensible to diverse groups of learners. This study presents an innovative approach to physics education through the integration of Large Language Models (LLMs) in the generation of interactive simulators developed in HTML and JavaScript. By employing targeted prompt engineering, teachers (even those with limited programming expertise) can co-design dynamic and customisable tools that model a wide range of physical systems, such as kinematics, wave propagation, electric or magnetic fields and energy transformations.

The work, conducted in collaboration with physics teachers and their students, demonstrates how AI-assisted simulation design supports active and inquiry-based learning, enabling learners to manipulate variables, observe real-time effects, and deepen conceptual understanding. Particular attention is given to the potential of these simulators to support students with special educational needs by offering simplified visualisations, multimodal representations, and adjustable levels of cognitive load.

Findings indicate that the integration of artificial intelligence into the development of digital learning resources constitutes a scalable and inclusive strategy for enhancing conceptual reasoning, promoting student engagement, and fostering personalised learning pathways in physics education. This approach offers promising directions for strengthening STEM teaching practices, particularly within blended, digitally enriched, or highly differentiated learning environments.

keywords: Vibe Coding, Artificial Intelligence, Physic simulator, IBL.

Refences: