Infant milk is a manufactured food designed for feeding babies and infants under 12 months of age. This product is indicated when, for some reason, the baby can not be fed by breast milk. It is used in case of intolerance to milk from farmed animals and is widely used for the nutrition of newborns in artificial lactation, in place of breastfeeding. It can be used as a complete or partial substitute for human milk. In the current study, the committed effective dose and lifetime cancer risk due to ingestion of 4 K, 226Ra, and 228Ra were estimated using activity concentration obtained from high-resolution gamma spectrometry. Consumption data and dose coefficients were obtained from Brazilian Institute of Geography and Statistics and data published by ICRP 119. Infant milk and water samples were collected in the Rio de Janeiro trade, placed in a polystyrene container, and then sealed to reach the secular radioactive equilibrium condition (at least 45 days). The samples were analyzed by gamma spectrometry using a High Purity Germanium (HPGE). Detection efficiency for each sample was obtained with LabSOCS software. For infant milk samples, the highest values of activity concentrations were (242 ± 9) Bq.kg−1 for 4 K, (5 ± 1) Bq.kg−1 for 228Ra and presented values below the detection limit for 226Ra. The higher values obtained for water samples were (10 ± 1) Bq.l−1, (5 ± 1) Bq.l−1, (5.3 ± 0.1) Bq.l−1 for 4 K, 226Ra and 228Ra, respectively. These values of committed effective dose are below the values stipulated by UNSCEAR (290 μSv.y−1) and by the ICRP (1 mSv.y−1). The values obtained for lifetime cancer risk are considered negligible, once they are below the limit considered for taking action (10−4). It can be concluded that the values calculated do not present a significant risk regarding the intake of infant milk in the first two years of the child's life.