The Office of the National Data Commissioner (ONDC) has been established to enable access to data held by government, accompanied by legislation for improved data access along with a data governance framework. The approach to and conceptualisation of ‘data governance’ and of public and government data in this framework form the focus of this paper. At the heart of this regulatory endeavour are concerns with (data) sensitivity, access and openness. A ‘Five Safes Framework’ has been developed to provide a basis for the release of government data and likely other sources (including research and commercial data) and accelerate access to sensitive data across the social sciences and related disciplines. This framework conceptualises safe distribution of sensitive data according to five elements: people, projects, data, settings and outputs. Not only do the constituent elements of the framework intersect in a range of ways, they are also interpreted differently by people according to the contexts in which they work. This paper outlines a preliminary analysis of this governance framework, and its expression of what data is and does and how it should it be mobilised. It forms part of a larger program of work funded by the Australian Research Data Commons: the CADRE project – Coordinated Access for Data, Research and Environments – which aims to establish a shared and distributed sensitive data access management platform for the social sciences and related disciplines, to enable data owners and users to address core concerns around governance, creation, management and access of sensitive data for research. We argue that this solution-focussed aspiration tends to compress matters that are at once technical, administrative, regulatory, ethical and epistemological. In this paper we report on our work interrogating the Five Safes and analysing how this Framework operates for researchers working across quantitative and qualitative data, the different ways in which the Framework can be interpreted and applied, and the diverse and unintended effects it might have on researchers and knowledge creation – with specific reference to implications for research and policy in the field of education.