Abstract Human activities are transforming landscapes and altering the structure and functioning of ecosystems worldwide and often result in sharp contrasts between human‐dominated landscapes and adjacent natural habitats that lead to the creation of hard edges and artificial boundaries. The configuration of these boundaries could influence local biotic interactions and animal behaviours. Here, we investigate whether boundaries of different degrees of ‘hardness’ affect space utilization by migratory species in Serengeti National Park, Tanzania. We deployed camera traps along transects perpendicular to the national park boundary at three different locales. The transects were located in areas that consisted of two types of human–wildlife interface: a sudden transition from the national park into agro‐pastoral land use (termed a ‘hard’ boundary) and a more gradual transition mediated by a shared usage area (termed a ‘soft’ boundary). Camera traps were placed at 2 km intervals along each 10 km transect from the edge towards the core of the park and were programmed to collect images hourly between dawn and dusk between June 2016 and March 2019. We used a deep neural network to detect the presence of wildlife within images and then used a Bayesian model with diffuse priors to estimate parameters of a generalized linear model with a Bernoulli likelihood. We explored the binomial probability of either wildebeest or zebra presence as a function of distance to the boundary, the rate of grass greening or drying (dNDVI) and the concentration of grass protein. There was a strong negative effect of distance to boundary on the probability of detecting wildebeest or zebra; however, this was only observed where the transition from human‐dominated landscape to protected areas was sudden. Conversely, soft boundaries had little to no effect on the probability of detecting wildebeest or zebra. The results suggest that boundary type affects migratory species occurrence. The implications of these findings suggest that hard boundaries reduce the effective size of conservation areas; for many species, the area used by wildlife is likely less than the gazetted area under protection. The impacts may be severe especially for narrow protected areas or dispersal corridors.