Many family doctors and residents in primary care are expected to participate in research but they have limited time. At the same time, generative artificial intelligence (AI) tools are rapidly entering clinical and academic settings, often without clear guidance on how to use them critically and safely. This workshop addresses the gap between research training needs in primary care and the unstructured use of generative AI by offering a structured, practice-based approach to protocol development that integrates methodological rigor, ethical reflection and equity considerations.By the end of the workshop, participants will be able to: a) Formulate relevant primary care research questions and objectives with structured support from generative AI. b) Use AI tools to draft and refine key sections of a research protocol (background, methods, variables and analysis plan). c) Design effective prompts, verify AI outputs and detect common errors and biases. d) Recognise ethical, confidentiality and equity issues linked to the use of AI in health research.The workshop combines a brief introductory talk, live demonstrations of generative AI tools, small-group work based on real primary care problems, and structured group discussion. Participants will work on their own or shared project ideas, using AI to generate and improve protocol components while critically appraising the outputs. Facilitators act as guides and methodological “co-pilots”, emphasising critical thinking rather than passive use of technology.Participants will leave with a structured outline of a research protocol relevant to their own context and a small, reusable “prompt bank” to support future projects. They will also gain practical skills in critical, safe and equity-aware use of generative AI.This workshop offers a novel combination of hands-on protocol development, critical AI literacy and an explicit focus on ethics and inequalities, supporting primary care professionals to strengthen their research capacity without relinquishing professional judgement or methodological quality.