Adaptive learning systems typically involve learner profiles with sensitive information and activities that might breach learner's privacy, such as tracking, profiling or collaborative filtering. In this paper, we discuss privacy aspects of personalisation and adaptation in smart spaces for learning. A smart space for learning is a system for mediation of learning services, which is being developed in the EU IST ELENA project. We present privacy threats and requirements in smart spaces for learning with focus on a learner, we model learner profile data that are necessary for privacy protection, and propose privacy-enhancing technology based solutions.