Abstract
OBJECTIVES: to provide evidence on the usage of artificial intelligence in Brazilian Primary Health Care. METHODS: scoping review, conducted on February 2025. Searches were accomplished in six databases. Selection process was conducted by three reviewers. Data were extracted using a form by two reviewers. Results were synthetized in a narrative way. RESULTS: 27 studies were included out of 981 retrieved. Machine Learning and Deep Learning were used in the studies to: make estimates and predictions, track and diagnose health conditions and complications and analyze determinants of universal health coverage. The main strengths found for using them were: to optimize resources; to provide early diagnosis; to allow risk stratification; to facilitates access. The main challenges for using them were: poor access to treatment and data availability. CONCLUSIONS: although the strengths of AI applications in Brazilian healthcare system were highlighted, its implementation should address the guarantee of the effective right to health.