Media reports propose a time when artificial intelligence (AI) programs will supplement or replace doctors. Whilst this may seem outrageous, programs with AI capacities are already embedded in the clinical workflow of a digitally enabled practice. For example, warnings when contraindicated medications are prescribed utilises simple if-then logic and natural language processing (NLP) software. Similarly, clinical decision support programs can prompt doctors to consider ordering a test following diagnosis to improve care. These programs could potentially reduce the family doctor’s cognitive burden and improve patient diagnosis or early intervention. We aim to review NLP models in primary care since 2020.A research protocol for this review has been developed a priori. MESH and free-text search terms were developed for the following 3 concepts – NLP models, general practice, risk prediction or classification. A literature search conducted on 5 databases (Pubmed, Medline Ovid, Scopus, Web of Science, CINAHL) was conducted in July 2025. 695 non-duplicate articles were identified and uploaded to an online literature management software (Covidence). Titles and abstracts were screened according to pre-defined inclusion criteria by two authors. Discrepant findings were independently adjudicated by a 3rd reviewer. Following screening, exclusions were made based on the pre-determined criteria. Included papers were reviewed and analysed using Nvivo software.There is no up to date published review of current applications of NLP for risk prediction in primary care. This scoping review fills this gap. A review of all AI implementation studies in 2020 reported on 90 studies relating to community based primary health care. Of these, only 3 utilised NLP models for risk prediction.Potential benefits to patients, practitioners and the healthcare system have been reported in a systematic review of AI implementation studies in community based primary health care.Applications utilising NLP that can be used to improve risk prediction from free text clinical notes, potentially enhancing existing clinical decision support systems and future applications in primary care, however their use remains in early stages.The digital health revolution has long been progressing, however, as wielders of evidence-based practice, we must assess the reality of current applications of NLP models