Broad-consent models for human studies, which leave decisions on data-sharing to the researchers, may not be appropriate for work with indigenous peoples. Making the sharing of data almost impossible is also problematic. Everyone stands to benefit from responsible data-sharing innovations that can be applied more widely.

The largest volume of genomic data amassed so far from Aboriginal Australians is deposited in the European Genome–phenome Archive (A.-S. Malaspinas et al. Nature 538, 207–214; 2016). The Australian board for ethical review permits access to the data only for “verification” purposes, and subject to ethical approval. For any other use of the data, researchers must seek permission from the original researchers, obtain ethical approval and contact the donors individually for their re-consent. This becomes prohibitive because donors are spread all over the continent.

These hurdles may be intended to protect participants from the harms of data-sharing, but they also exclude them from the potential benefits. Other models better address the foundational principles of responsible sharing, as proposed by the Global Alliance for Genomics and Health (B. M. Knoppers HUGO J. 8, 3; 2014), and the control of research by indigenous study participants. The National Centre for Indigenous Genomics (see go.nature.com/2slp6q6) in Canberra, for example, combines indigenous governance with dynamic consent that allows project participants to opt in or out of having their data shared (J. Kaye et al. Eur. J. Hum. Genet. 23, 141–146; 2015).