Ambient Intelligence constitutes a new paradigm for the interactions among intelligent devices and smart objects acting on behalf of humans. The ultimate goal of Ambient Intelligence systems is to transform our living and working environments into “intelligent spaces” able to adapt to changes in context and to users’ social needs. Such systems must have the capability of interpreting their contexts.Moreover, the open nature of Ambient Intelligence systems and the unnoticeable ways in which various sensors may access users’ personal data make two seemingly antagonistic requirements: preserve users’ privacy and facilitate convivial interactions among humans and devices. The concept of conviviality has recently been introduced as a social science concept for Ambient Intelligence to highlight soft qualitative requirements like user friendliness of systems. To make Ambient Intelligent systems desirable for humans, they must be convivial and private, i.e., fulfil both their need for social interactions, making communication and cooperation among participants easier, while also preserving the privacy of each individual. Hence the need for trade-offs between being private and convivial; intuitively the former keeping information to a small circle of insiders while the later shares it with all. The aims are to enable knowledge sharing for the collective achievement of common objectives among entities through coalition formation, while at the same time respecting each entity’s privacy needs. Hence, the CoPAInS project is about designing and developing models of interaction among intelligent devices; it is also about their corresponding methods. The main goal of CoPAInS is to facilitate privacy and conviviality in Ambient Intelligence and to provide tools for it. To achieve our objective, first we develop a formal framework for context representation, privacy and conviviality, using methods of Knowledge Representation & Reasoning. Second, we validate and verify privacy and conviviality properties of Ambient Intelligence systems using model-checking techniques. We then simulate and test use cases from the field of Ambient Assisted Living. Third, to demonstrate the feasibility of our approach and its usefulness to support Luxembourg citizens’ everyday activities, we apply the gained expertise to use cases provided by Hotcity. Finally, we produce guidelines to assist system designers in the development of privacy-aware and convivial “assistive systems”.