b24a7ca1d6f670fa

KRYPTOS_SUMMARY.PDF

NSA·UFO_FOIA·pdf·11.7 MB·26 pages

Scores

3.5
Document value
0.0
Cross-references
6.0
Provenance
4.4
Info density
4.0
Topic relevance
0.0
Anomalousness

OCR'd text preview (8 of 26 pages)

Source: embedded

page 0
DOCID: 4145037 
CONFIDENTIALHREL TO USA, FVEYH20320108 
(U) THE CIA KRYPTOS SCULPTURE: 
A summary of previous work and new 
revelations in working toward 
its complete solution 
by 
PL 86-36/50 USC 3605 
CO~:FUltJ;NTIALh'IU;L TO USA, FYEY//20320108 
1 
pprnved 
or Release 
v ~-~SA on 
9-16-2014. 
OIA Case 
.. 65414 
page 1
DOCID: 4145037 
CONFIBENTIALHREL T0 USA, FYEYH26326168 
(U/ /FOUO) Back in 1990, the Central Intelligence Agency's Fine Arts Commission canvassed 
Washington, DC, area artists, asking for proposals for costs and designs for a sculpture which 
would be erected in the CIA's new Courtyard outside its cafeteria. The winning sculptor was 
James Sanborn, and he created a nice little puzzle. It stands about 3 metres (10 feet) tall, and con-
sists of two curved copper plates standing side-by-side, forming an "S" shape if viewed from 
above. 
(U/IFOUC>) Before we look at the cipher on the sculpture, he
page 2
4145037 * + DOCID ABCDEFGHIJKLMNOPQRSTUVWXYZABCD AKRYPTOSABCDEFGHIJItLULMNOQOUVWKZKRYP BRYPTOSABCDEFGHIJJILMNQUVWXZKRYPT CYPTOSABCDEFGHIJLUMNQUVWXZKRYPTO DPTOSABCDEFGHIJILMNOQOUVWX2ZKRYPTOS ETOSABCDEFGHIJLUMNQUVWXZKRYPTOSA (U/FFOUCOSF ig. 1. Vigenere Square from KRYPTOS sculpture FOSABCDEFGHIJLMNQUVWXZKRYPTOSAB GSABCDEFGHIJLMNQUVWXZKRYPTOSABC HABCDEFGHIJULUMNQUVWXZKRYPTOSABCD -IBCDEFGHIJLMNQUVWXZKRYPTOSABCDE TCDEFGHIJLMNQUVWXZKRYPTOSABCDEF KDEFGHIUJLMNOUVWXZKRYPTOSABCDEFG LEFGHIJLMNQUVWXZKRYPTOSABCDEFGH MFGHIJLMNQUVWXZKRYPTOSABCDEFGHI NGHIUJULMNQUVWXZKRYPTOSABCDEFGHIJL OHITJTLMNQUVWXZKRYPTOSAB
page 3
4145037 * * DOCID (U//F6%8} This process continues through the rest of the 26 lines labeled A-Z at the beginning of each line, except for the line labeled “N.” On this line, the shifted KRYPTOS sequence begins .CDEF..., but instead of following with the first four letters from that offset, GHTIJ, five letters appear instead, GHIJL. GHIJ..., and finishes .. (U/2666} The 28th line of the Vigenere Square is just a repeat of the top line, the A- Z alphabet followed by the letters ABCD. So we have two things to notice from this side of the sculpture: we have the Direct Standard alphabet (A- Z), and
page 4
DOCID: 4145037 CONFIDENTIAL REL-FO-US A, FVE¥/20320108— (U//F6G6> There are a few other things to point out here as well. The 15th line of the cipher begins with the letters END, and of the letters YAHR which follow it, the Y, A, and R are raised up about | cm. (about 1/2 inch). This made analysts at NSA wonder is the END was perhaps the end of a certain portion of cipher. And maybe the Y, A and R floating in the air a bit higher than the other letters on that line may signify something else as well. (U//FO86} Well, in February of 1999, a CIA employee named David Stein, stated to the world tha
page 5
DOCID: 4145037 -CONFEIDENFEAEHREE-FO-ES A, EVE ¥#203264-08— (U/fFOUOS This particular individual started with the beginning of the cipher itself, EMUFPH- ZLR. He happened to notice the KRYPTOS alphabet in the Vigenere Square, and so he tried a very reasonable thing, simple addition. E is the Sth letter of the alphabet, and K is the 11th letter of the alphabet, so if we add 5 and 11 we-get 16, and P is the 16th letter of the alphabet. Then M, the 13th letter, is added to R, the 18th letter. That adds up to 33, but in mod26 arithmetic, that’s 5, which is an E. The same process continues until PE
page 6
(b+.,.()J 
\b+.'('9'JC~DJ5c ~:J.±~;>037 
PL 86-36/50 USC 3605 
CONFIBENTIAL/-/REL TO USA, FYE¥f/29329198 
CONFII>RNJ'IAL/IREL J'O USA, FVEYH20329198 
7 
page 7
-DOCID: 'RPWeEwWdns Pp 4145037 “CONFIDENTIALAREE FC-ESA, FVEY7/20320408- (U/7FOUSG} The strength in polyalphabetic substitution is that the same plaintext letter will not always be enciphered to the came cipher letter.as each encipherment is dependent upon the alpha- bet enciphering it. And with multiple alphabets, there are multiple ways for each plaintext value to change. Similarly, identical ciphertext letters did not always evolve from the same plaintext value: they can be spawned from two entirely different plaintext values. (U/FO0) The weakness of polyalphabetic substitution is if you ha

Full text and original imagery available on Internet Archive →