In relation to clinical trials, it is far more usual to speak of the community (singular, static) than of publics (multiple, emergent). Rarely defined, the community is commonly taken to be the existing people in a given area, which the trial will engage, mobilise or sensitise to facilitate successful recruitment and retention. Communities are assumed to pre-exist the research, to be timeless, and to be a whole (sometimes consisting of different parts, referred to as stakeholder groups). In this paper, we suggest a conceptual shift from ‘trial community’ to ‘experimental publics’. Using an empirical case study of an HIV prevention trial in Zambia, we draw out the following key points: firstly, publics do not pre-exist research activities but are enacted in concert with them. Secondly, publics are dynamic and transient. And thirdly, experimental publics are situated at the intersection of various forms of inclusion and exclusion, both locally and globally. Our findings emphasise the need to create long-term forms of participation in science, which transcend both the instrumental goals and the individual timelines of specific trials.