Government data review | November 14, 2026
A UFO percentage is useful only when its denominator and categories are clear. France’s GEIPAN gives readers both, and its November snapshot shows why an unexplained-case count should never be treated as a count of extraordinary vehicles.
The dated snapshot
The official statistics page, dated November 10, lists 3,404 classified, published cases across observation years: 955 in A, 1,323 in B, 1,020 in C and 106 in D. The displayed percentages are 28.1, 38.9, 30.0 and 3.1 respectively. These rounded figures total 100.1 percent; the case counts sum correctly. This is a cumulative archive snapshot, not a count of sightings during November.
The same page’s breakdown by identified phenomenon uses a narrower sample of cases published in 2023–2025, assessed as of June 1, 2025. Readers should not silently apply that chart’s proportions to every case in the larger archive.
Two different reasons for an unidentified result
GEIPAN’s classification guidance defines A as identified and B as probably identified. C means the information is insufficient; D means the phenomenon remains unidentified after investigation. The agency allows classifications to change when new information arrives, and says D generally follows a field investigation involving the witnesses.
That distinction is consequential. Combining C and D into one dramatic total would erase whether investigators had enough material to test an explanation. Conversely, treating every C case as already explained would also misrepresent the record. Neither label supplies an origin, propulsion mechanism or species.
What a useful comparison would require
Our assessment is that an archive trend needs dated snapshots and a record of reclassifications. A change in the percentage of D cases could reflect newly published cases, revised old investigations or a changing denominator. It cannot, by itself, establish a change in unusual activity over France. We did not independently audit all 3,404 case files or reconstruct the archive’s historical revisions.
NASA’s UAP study framework supplies an independent methodological comparison: it asks what data already exist, what additional observations should be collected and which analysis methods can assess them. NASA did not verify these French counts. The two GEIPAN pages are one institutional source chain; NASA supports the measurement approach rather than the individual classifications.
What to watch: publishable case-level evidence and documented reclassifications would be more informative than a headline percentage alone. Keeping incomplete cases separate from investigated unknowns makes that work easier to evaluate.
Hero image: original AI-generated conceptual artwork. The observatory, folders and displays are fictional and do not depict GEIPAN facilities or case evidence.