Project Blue Book Special Report No. 14

In 1955 the U.S. Air Force published the single largest statistical study of UFO reports it ever produced. Its narrative said the unexplained cases would disappear with better data. Its own numbers set aside 434 sightings as "UNKNOWN" and ran statistical tests showing those unknowns were measurably different from the explained ones. That tension is why Special Report No. 14 is still argued over - and the document is now part of the Trump PURSUE disclosure.

What it is

The document's own title page reads "Project Blue Book Special Report No. 14 (Analysis of Reports of Unidentified Aerial Objects)," Project No. 10073, dated 5 May 1955, issued by the Air Technical Intelligence Center at Wright-Patterson Air Force Base, Ohio (the copy in this release is marked "Copy No. 35," "For Official Use Only"). Per the report, the study was "prepared by a panel of scientists both in and out of the Air Force." It is widely documented in the public record that the statistical work was carried out for the Air Force by the Battelle Memorial Institute; that contractor attribution is established external context, not a claim drawn from this file.

Unlike a single sighting report, Special Report No. 14 is a data project: it took the accumulated body of Air Force UFO reports, reduced them to coded "work sheets" and punch-card form, and ran statistical comparisons between the cases that could be identified and the cases that could not.

The numbers, from the document itself

Per the report's own figures, of the 2,199 "object sightings," 434 - 19.7% - were evaluated as UNKNOWN (Figure 3 in the document), a share comparable to the 21.8% it attributed to astronomical causes. The report then ran chi-square statistical tests of "Knowns versus Unknowns" on the basis of color, number, shape, duration of observation, speed, and light brightness - in effect, testing whether the unidentified sightings looked statistically like misidentified known objects, or like a distinct population.

The report also tracked how the unknown rate fell as investigation improved: among the 854 cases received in 1953-1954 the unknown share was 9%, and after a revised "on the spot" investigation program (AF Reg. 200-2, carried out by the 4602d Air Intelligence Service Squadron) the unknown rate for early-1955 reports dropped to 3%.

What it concluded - and the tension

The report's narrative conclusion was deflationary. In its own words, "it is believed that all the unidentified aerial objects could have been explained if more complete observational data had been available," and as investigation methods improved "the number of unexplained cases has decreased rapidly towards the vanishing point."

That is the tension at the heart of Special Report No. 14, and it is visible in the file itself: the prose argues the unknowns are an artifact of incomplete data, while the study simultaneously retains 434 sightings (19.7% of object sightings) as UNKNOWN and runs statistical tests treating them as a category worth distinguishing. Both readings are cited to this day - one side reads the conclusion, the other reads the tables.

What PURSUE adds

Special Report No. 14 is not new to the public record - it has long been available, including on the CIA's own reading room. Per this file's released summary, the PURSUE copy is the USAF report behind a CIA "Official Record Copy" cover sheet, and "with the exception of the handwritten note on the first page, the content of this document has been available on CIA's public website." So what the disclosure adds here is provenance: the document in the canonical war.gov PURSUE set, carrying a 2026 release stamp, mirrored on this site with a SHA-256 hash so you can confirm the copy you are reading matches the government's. This file sits in the broader CIA UFO files cluster (19 files).

What Special Report No. 14 does NOT establish

• It is not evidence that UFOs are extraterrestrial. The 19.7% "UNKNOWN" classification means exactly that - unidentified for lack of sufficient data - not "confirmed as anything in particular."
• Its headline conclusion is the opposite of a smoking gun: the report argued the unexplained cases would shrink "towards the vanishing point" with better investigation, and documented that they did fall (to 3%) under improved reporting.
• It is a 1955 statistical analysis of mid-century reports. It is a primary source on how the Air Force measured and characterized the data, not a verdict on any individual sighting.

How to verify everything on this page

1. The file card above links to CIA-UAP-015's page on this site, where you can view the document, download the original (SHA-256 90e05e77..., verified against war.gov), and click through to the war.gov source URL.
2. The figures cited here - 3,201 total sightings, 2,199 object sightings, 434 / 19.7% UNKNOWN, the 1953-54 (9%) and 1955 (3%) unknown rates, the chi-square "Knowns versus Unknowns" tests, and the "vanishing point" conclusion - are all taken directly from the text of the released document, not from outside accounts.
3. The "Battelle Memorial Institute" contractor attribution is long-established public-record context about Special Report No. 14, provided here as background; the file itself states only that it was "prepared by a panel of scientists both in and out of the Air Force."

Bottom line

Special Report No. 14 is the government's most-cited UFO statistics document precisely because it can be read two ways. Its conclusion waves the subject away; its data quietly keeps a fifth of the analyzed sightings in the "UNKNOWN" column and runs tests on them. Now you can read the actual report behind both readings, verified against the source.