Artificial intelligence in physics education: A systematic review of content coverage, implementation models, learning impact, and pedagogical challenges
Abstract
Artificial intelligence (AI) has made a fast entry into the world of physics education research, but the empirical literature is still fragmented according to content domain, pedagogical function, and evaluation criteria. Leveraging disciplinary epistemic practices, instructional scaffolding theory, and self-regulated learning, this is a systematic literature review of 44 empirical studies published from 2021 to 2025 to examine the implementation of AI in physics education, where it is concentrated, the learning functions it serves, and the epistemic risks associated with its use. Following the guidelines of PRISMA, the review combines frequency analysis and thematic synthesis in order to search for patterns beyond isolated outcomes across studies. The results show that there is a high concentration of AI applications in the areas of mechanics, thermodynamics, and modern physics, which are characterized by machine-interpretable problem structures and high demands for abstraction. AI is mainly applied in conversational tutoring, automated feedback, and personalization systems, and the benefits have been shown to be focused on short-term learning performance, engagement, and instructional efficiency. However, common themes among the contexts are observed, such as epistemic unreliability, overreliance on AI outputs, lack of AI literacy, and fragility of methodology, which directly pose a threat to disciplinary reasoning practices that are central to learning physics. Far from…
How strong is the evidence?
Evidence Hub reads this as a systematic review. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Chaos and quantum thermalization
- The Dalton quantum chemistry program system
- Cord Blood Leptin and Adiponectin as Predictors of Adiposity in Children at 3 Years of Age: A Prospective Cohort Study
- Quantitative metabolomics by H-NMR and LC-MS/MS confirms altered metabolic pathways in diabetes
- Stable adiabatic quantum batteries
- A novel self-lipid antigen targets human T cells against CD1c(+) leukemias
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.