Incident: 22 June 2025, Paris. Investigation: GEIPAN report dated 27 November 2025. Outcome: category A, identified as one or more windborne balloons. The source of the balloon or balloons was not established.
A resident watching from a terrace described a dark appearance whose shape seemed to change between a single body and linked rounded forms. It moved across the sky and appeared to pause. A second person saw the end of the episode, but the file contains one principal submitted account.
The camera corrected the clock
The investigation compared four photographs and a short video. Image metadata put the photographs at 21:48–21:50, about an hour later than the initial recollection. The witness confirmed that the camera times were correct. This matters because weather, lighting and aircraft checks must use the right time.
The correction does not invalidate the experience. It shows why a contemporaneous file timestamp and a remembered clock reading need to be distinguished before drawing conclusions. UKPX retains the corrected evening timing rather than publishing both versions as if they were two sightings.
The explanation and an unsuccessful lead
GEIPAN found the appearance and apparent path consistent with windborne balloon material. Its report checked a balloon exhibition at the Grand Palais; the organisers denied that any installation had escaped at that time. The final category A finding did not establish a particular release point.
That unsuccessful enquiry is worth retaining. An identified class of object does not necessarily come with an identified owner. The exhibition should not be presented as the proven origin just because it was investigated.
Why a pause is not proof of control
Apparent stillness is not, by itself, a demonstration of deliberate hovering. Viewing geometry and changing local air currents can affect what an observer sees. In this case the agency considered the pause within its windborne-object explanation rather than treating it as an independently measured manoeuvre.
The complete report is the relevant source for the agency’s conclusion. The images and witness estimates do not provide UKPX with an independently measured altitude, physical size or flight-control system. The case is retained as an explained 2025 observation, with the unresolved launch provenance kept separate from identification of the likely object class.
Map and source handling
The map uses a broad city reference. No private terrace is pinpointed and no directions to a witness’s home are given. The linked report contains third-party photographs and maps; UKPX links it rather than assuming that every image inside it is free to copy.
Sources and record status
Primary case source · GEIPAN investigation report, 27 November 2025. Sources checked 10 November 2026. Reported observations, source assessments and UKPX’s discussion of evidence limits are distinguished above. No independent identification is claimed beyond the stated source finding.
Map reference: Broad Paris city reference rounded from the geolocated contextual photograph; not a witness terrace, private address or balloon launch point. Coordinate source. Incident year: 2025.