Recent developments around the Smart Grid promise more efficient power generation and distribution. In contrast to the current design of the electricity grid, the Smart Grid is heavily based on IT for managing communication flows. Given the large number of market participants, data exchanged should be marked up in a way that facilitates flexible modelling, extension and integration. In addition, since much of the data exchanged is highly sensitive - comprising e.g. consumption data - current organisational means for privacy enforcement are not sufficient. In this paper, we propose a novel framework, which addresses these privacy issues by means of a machine-interpretable representation of the user intent. More precisely, such machineinterpretable specifications enable an automated access control and enforcement of privacy rights on a technical level.