Purpose: The objective of this project is to develop a decision support tool based on an artificial intelligence technique (Bayesian network) that predicts breast cancer that was initially applied to a group of Colombian women. Methods: We developed a Bayesian network consisting on five nodes that includes current age and three risk factors for breast cancer in Colombia (age at menarche, age at first live birth and first and second degree relatives with breast cancer). The Bayesian network was developed using information retrieved from interviews of 214 patients with breast cancer in Bogotá Colombia. A database was designed for the recollection of data for future probabilities calculation. All subjects interviewed were women with breast cancer treated at Fundación Santafé de Bogotá. Conditional probabilities were calculated using the prevalences of each risk factor and population age for Bogotá. This demographic data was obtained from Profamilia, a non-profit organization that provides family planning services. The preliminary outcomes of the model were validated by Colombian experts (medical doctors). The next phase of the project is aimed at applying the tool developed to a representative population of women of Bogotá and including more risk factors. This will allow us to discover what are the principal risk factors for this specific population and be able to improve diagnostic analysis and treatment. Results: Having first and/or second degree relatives with breast cancer was found to have the greatest impact in risk for breast cancer. It was found that a woman with a first degree relative and no other risk factor has a 17.4% chance of developing breast cancer compared with an 8% chance of a woman with no risk factors. The age at menarche was found to influence in second place, followed by the age at the first live birth. It must be pointed out that these results apply for the pilot study population. Conclusions: Our Bayesian network is the first step in an ambitious project which intends to perform valid calculation of breast cancer risk for all Colombian women. This being said, the tool is useful as it is as a decision support system for doctors willing to use it so that the treatment followed is more accurate and methodologically supported.