Abstract: There is considerable uncertainty surrounding population estimates for Aboriginal and Torres Strait Islander Australians at both the aggregate and subregional levels. Given well-documented limitations of efforts to enumerate the existing population at a point in time, the task of projecting future populations is even more problematic, and currently no regular projections are produced below the state level. However, information on future trends in the Aboriginal and Torres Strait Islander populations at more disaggregated geographical and demographic levels would be of enormous value for policy and planning. This is particularly so for remote Australia and the remote Aboriginal and Torres Strait Islander communities for which there has been ongoing controversy about the provision of services and infrastructure and even their very ‘viability’. Policies, planning and funding models based on inaccurate assumptions of future populations are likely to lead to misallocation of resources and sub-optimal outcomes for Aboriginal and Torres Strait Islander people. This paper sets out a proposed methodology for generating projections of the Aboriginal and Torres Strait Islander populations in Outer regional and Remote Australia, down to the community (or ‘Indigenous Location’) level. Based on 2006 and 2011 Census data, projections for remote Aboriginal and Torres Strait Islander communities for 2016 are presented by age group and gender, and the implications of those estimates for policy are discussed. These estimates should be considered experimental and are selected from a range of models under development. Release of 2016 Census data will enable validation of the different models and provide a panel dimension that can be used to substantially enhance modelling for generating 2021 projections and beyond.