Credibility Assessment, truth verification or lie detection – many terms point to the perpetual endeavour of developing procedures that are supposed to ensure the authenticity of others. This is especially true when the place of action is a vulnerable infrastructural site and the target action is to avert threats. Airports are large architectural filtering systems, in which people and goods are channelled through technological and human screening procedures. But airports also serve as a prominent place for imaginaries of automated control technologies. Although at least in Germany, such technologies are still subject to laboratory work and field trials, they share the notion of searching for harmful intentions with current filtering systems. Embedded in the nexus of STS scholarship, critical security and surveillance studies, the article sheds light on the differences and commonalities of human-centred suspicion practices at the airport and its envisioned automated equivalents in the laboratory. Based on insights of focused ethnographic research at German airports and in-depth interviews with researchers, the article presents three empirical case vignettes, highlighting the rationales of human border control, airport security screeners and researchers. Focusing on the role of tacit knowledge for identifying indicators for harmful intentions, the human-centred rationale of baselining is put into relation to the so-called ground truthing in computational contexts. The field insights suggest that, despite their differences, each mode of thinking and doing relies on conceptualising a so-called phenomenon of harmful intentions as a non-verbal physical display, that manifests itself as an epistemological object as well as a single truth.