In the last decades, Conversation Analysis has experienced an “embodied turn” (Nevile 2015): thanks to technological advancements, naturalistic video data have become more and more available, and the research focus has moved from verbal to multimodal interactions. As a result, researchers are nowadays faced to the use of video data, to their possible applications for teaching, professional training and research’s popularisation. The ICAR Lab has a long tradition in the study of social interaction and in collecting video corpora in naturally occurring settings. Through a large variety of projects, it has shared answers to methodological and deontological questions concerning fieldwork and technical aspects of video recording (see Jouin-Chardon et al. 2010). More recently, our research group produced short films for promoting our research activity and raising awareness about societal issues (Piccoli & Ursi 2015, DISAL project, REMILAS project). This contribution will relate our experience in film creation for such purposes. In a first project, three short films issued from three different corpora were produced. Initially designed as a feedback for participants, in an attempt to improve mutual professional relationship (see Baude et al. 2006), they also turned out to be a useful tool to teach the interactional approach and to promote research to a widespread audience. In another still on-going research on multilingual communication between healthcare providers and migrants in France (the REMILAS project), video data are used not only for scientific purposes, but also for healthcare professionals and interpreters’ training, as well as artistic initiatives aimed at promoting a more informed and active citizen participation to nowadays society.