Hall XII · Unsolved · CIA Courtyard Modern · 1990 K1–K3 Broken · K4 Unsolved

Kryptos

CIA courtyard sculpture, 1990 — three sections solved, one (K4) still open

ArtistJim Sanborn
Cryptographic consultantEdward Scheidt (former CIA)
InstalledNovember 3, 1990 · CIA HQ, Langley
Copper screen1,735 letters cut into the main screen
Full artworkCourtyard screen + entrance Morse / compass installation
K1Vigenère, key PALIMPSEST · solved 1999
K2Vigenère, key ABSCISSA · solved 1999
K3Transposition · solved 1999
K4Unknown system · 97 characters · unsolved

Kryptos Expansion Lab

The sculpture is larger than K1-K4 alone. Explore the entrance Morse panels, the K4 clue windows, and the full installation layout.

K0 Morse Decoder

Load a phrase from the entrance granite slabs, hear the Morse at a selectable speed, and inspect its decoded text. These fragments are part of the larger Kryptos environment, not a separate puzzle by accident.

K4 Clue Visualizer

Sanborn's released cribs occupy only small windows of the 97-character K4 text. Hover any character to see whether that position is known or still open.

Installation Map

Kryptos is a two-part installation: the famous copper screen in the courtyard and the entrance granite / Morse / compass elements visitors pass on the way in.

Entrance Walk K0 Granite slabs Morse + compass Courtyard Plaza Copper screen Courtyard Lawn Sight lines toward the S-screen
Reading the layout

The entrance components establish the work's atmosphere before the visitor reaches the famous courtyard screen. That is why Round 3 treats Kryptos as an installation, not just a single panel of ciphertext.

The map distinguishes three zones: the entrance Morse / compass elements, the plaza where the 1,735-letter screen stands, and the surrounding courtyard space from which the sculpture is meant to be approached and read.

Why This Matters

Kryptos is a 12-foot copper sculpture installed in the CIA's courtyard in 1990, encoded with four separate cryptographic puzzles totaling 865 characters. It was meant to take the agency a few months to solve; instead it took the NSA seven years (they cracked K1–K3 first, in 1992, but kept it secret), and the public CIA team eight. The fourth section, K4 — only 97 characters long — remains unsolved more than 35 years later, despite hints from the artist Jim Sanborn (BERLIN, CLOCK, NORTHEAST, EAST) and two solving prizes that have gone unclaimed.

📜Historical Context

The CIA commissioned the sculpture from artist Jim Sanborn, who worked with retired CIA cryptographer Ed Scheidt to design four progressively harder ciphers. K1 used a modified Vigenère with the keyword PALIMPSEST and a custom KRYPTOS-keyed alphabet; K2 used the same system with key ABSCISSA. K3 was a transposition cipher quoting Howard Carter's description of opening Tutankhamun's tomb. K4 remains a mystery — Sanborn has revealed four "cribs" (BERLIN at letters 64–69, CLOCK at 70–74, NORTHEAST at 26–34, EAST at 22–25), but the system that generates the rest is still unknown. Many have tried; none have succeeded.

🗺️The Full Installation

The familiar copper S-screen is only half of Kryptos. The complete artwork spans two installation zones: the courtyard screen with K1-K4, and the entrance approach with red granite slabs, copper elements, International Morse code, a compass rose, and a lodestone. The CIA's own description treats these as one integrated work. That matters pedagogically: the puzzle is environmental. The visitor is meant to enter through a field of clues before reaching the famous ciphertext.

🔢1,735 Letters, Not 1,800

The authoritative count from the CIA is 1,735 alphabetic letters cut into the copper screen. The often-repeated 1,800 figure comes from workshop recollections about the approximate labor involved, not the final installed total. The discrepancy is worth noting because Kryptos lore has accumulated many rounded numbers and repeated anecdotes; this museum uses the CIA count when discussing the main screen itself.

📡K0 — The Morse Code Panels

The entrance granite slabs carry Morse fragments sometimes nicknamed K0 by solvers. Phrases such as VIRTUALLY INVISIBLE, SHADOW FORCES, LUCID MEMORY, and the famously garbled DIGETAL INTERPRETATIT are part of the installation's texture of partial legibility. Round 3 treats these panels as first-class Kryptos material because they expand the work from a four-passage challenge into a layered meditation on concealment, navigation, and interpretation.

🤝The Scheidt Partnership

Sanborn did not build Kryptos alone. He spent roughly four months working with retiring CIA cryptographer Edward Scheidt, who helped design the cipher systems and later described the challenge as 9 out of 10 in difficulty. Scheidt also emphasized that K4 changes methodology rather than merely increasing difficulty. That is one reason many solvers think K4 cannot be attacked by extending the techniques that break K1-K3.

🧮The Hill Cipher Conjecture

One of the more interesting published K4 theories comes from Bauer, Link, and Molle, who argued in Cryptologia that an anomalous extra L in the right-side tableau may be deliberate. With that extra letter in place, the letters HILL appear in sequence down the column, suggesting that K4 could involve Hill-cipher-style matrix encryption rather than another Vigenère-like method. This is only a conjecture, but it is stronger than casual internet speculation because it begins from a physical irregularity on the sculpture itself.

🧩K4 Clues Released Over Time

Sanborn has disclosed K4 in narrow slices rather than broad explanations: BERLIN (2010), CLOCK (2014), EAST (2020), and NORTHEAST (2020). In 2025 he clarified that the clock reference points specifically to the World Clock at Alexanderplatz in Berlin. These clues are enough to constrain local hypotheses, but not enough to expose the underlying system. That asymmetry is why K4 remains so compelling: the puzzle is neither untouched nor genuinely open-ended. It is deliberately half-lit.

✏️Passage 2 Correction

For years the accepted K2 ending looked like WESTIDBYROWS. In 2005-2006 Nicole Friedrich proposed the intended reading was WESTXLAYERTWO, and Sanborn confirmed that the sculpture omitted a needed S during transcription. The correction matters because it links K2 more clearly to Howard Carter's excavation language and strengthens the tomb / location thread running through K2, K3, and the speculation around K4.

📝Intentional Misspellings — or Workshop Drift?

Kryptos contains several famous oddities in its solved passages: IQLUSION in K1, UNDERGRUUND in K2, and DESPARATLY in K3. Sanborn has often described these as intentional artistic decisions. But later release of the working charts complicated that story: some irregularities appear to trace back to drafting or transcription mistakes made during the sculpture's fabrication. The right conclusion is not that one side is lying, but that Kryptos's "errors" are part design, part workshop history, and part legend.

ʸThe YAR Superscript

Three letters, YAR, appear in superscript near the beginning of the lower left section of the sculpture. They are the only superscript characters on Kryptos. No public explanation has ever been confirmed. Because the marker is so isolated, many solvers treat it as a deliberate pointer rather than decorative noise. The lesson is a good one for visitors: real cryptanalysis often begins with noticing exactly this kind of anomaly.

Who Is WW?

K2 asks, "WHO KNOWS THE EXACT LOCATION? ONLY WW." The leading conjecture is William Webster, Director of Central Intelligence at the sculpture's 1990 dedication. Sanborn has said he gave Webster a sealed envelope, but he has remained deliberately vague about whether that envelope held the full answer, only part of one, or simply ceremonial theater. The WW problem turns Kryptos into a question about institutional memory as much as cryptanalysis.

The Death Contingency

For years Sanborn said that, if he died before Kryptos was solved, he had arranged a way for correct solutions to be verified. The 2025 auction and Smithsonian sealing effectively replaced that older contingency with a different structure: partial archival closure, partial private ownership, and an unresolved question about who now has the authority to authenticate a solve. Kryptos therefore doubles as a case study in cryptographic governance.

🕰️2025: Sale and Seal

In 2025 the Kryptos story changed from an open public-art mystery into a partly archived, partly privatized saga. Sanborn announced his intention to auction material tied to the sculpture, confirmed the existence of a still-unrevealed K5, and requested that the Smithsonian seal some related files for 50 years, until 2075. The associated auction closed at $962,500. The cryptographic story is therefore no longer just about whether K4 can be solved. It is also about stewardship: who controls the answer to a puzzle embedded in public art once the artist can no longer manage that responsibility alone?

🎬Kryptos in Popular Culture

Kryptos escaped the CIA courtyard almost immediately and became one of the few real unsolved ciphers with mainstream cultural reach. Dan Brown hid Kryptos references in The Da Vinci Code jacket design and used it more directly in The Lost Symbol. Television series from Alias to The Recruit treat it as shorthand for elite cryptographic mystique. That afterlife matters: Kryptos is not just a cipher problem but a public mythology, which is why Round 3 also routes visitors toward Hall XIII: Cipher Culture.

⚙️How It Works

K1 and K2 use a Vigenère cipher built on a 26-letter "tableau" whose alphabet starts with KRYPTOS:

Plain  A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Tableau K R Y P T O S A B C D E F G H I J L M N Q U V W X Z
Each plaintext letter is shifted by the corresponding key letter using this alphabet (not the standard A–Z).

K3 is a complex transposition. K4 is unknown — possibly a one-time pad, a custom polyalphabetic, or a layered system Sanborn refuses to disclose.

💀How It Was Broken
K1, K2, K3 — broken 1992–1999
Complexity: Hard but solvable

NSA cryptanalysts Dennis McDaniel, Ken Miller, and others solved K1–K3 in 1992 using standard polyalphabetic and transposition techniques. The CIA's in-house team led by David Stein solved them independently in 1998. Computer scientist Jim Gillogly published the public solution in 1999. The plaintexts contain the words BETWEEN SUBTLE SHADING AND THE ABSENCE OF LIGHT… and a quotation about Tutankhamun's tomb.

K4 — open since 1990
Complexity: Unknown

K4 is only 97 characters long. Sanborn has provided four cribs to constrain the search, but the underlying system has resisted every public and (presumably) classified attempt. Speculation includes a one-time pad (in which case it cannot be solved without the pad), a customized periodic cipher with a very long key, or a multi-step composite. Sanborn has said he will reveal the answer if he dies, but as of 2026 it remains open.

🔬What It Teaches Modern Cryptography
Concept from KryptosModern Evolution
Custom-alphabet polyalphabeticsResistance to standard tableaus
Public challenge as research driverBitcoin puzzles, Project Euler, capture-the-flag culture
Short ciphertext is hardestStatistical attacks need data; 97 chars is too few
Quick Facts
HallXII · Unsolved
EraModern · 1990
SecurityK1–K3 Broken · K4 Unsolved
ArtistJim Sanborn
Cryptographic consultantEdward Scheidt (former CIA)
InstalledNovember 3, 1990 · CIA HQ, Langley
Main screen1,735 letters
Full artworkScreen + entrance installation
K1Vigenère, key PALIMPSEST · solved 1999
K2Vigenère, key ABSCISSA · solved 1999
K3Transposition · solved 1999
K4Unknown system · 97 characters · unsolved
🗿Sanborn's Other Cryptographic Sculptures

Cyrillic Projector extends Sanborn's cipher work into Russian text and was solved in 2003 by an international team that included Elonka Dunin. Antipodes places fragments of the Kryptos and Cyrillic Projector texts in conversation across two faces of one sculpture. Together they show that Kryptos is not an isolated stunt; it sits inside a broader body of cryptographic public art.

🏁Prior Public Solvers

K1-K3 were solved independently by multiple groups before the public realized it: NSA analysts in 1992-1993, CIA analyst David Stein in 1998, and Jim Gillogly publicly in 1999. The public story long centered on Gillogly because the classified teams could not discuss their work. Later FOIA efforts, especially by Elonka Dunin, widened that history.

🍽️The 2019 Kryptos Dinner

By 2019 Kryptos had developed its own solver culture. At invitation-only Kryptos dinners, attendees could reportedly submit K4 candidates directly to Sanborn. Those reports matter because they blur the line between public puzzle and curated community. They also frame the later 2025 discovery controversy: whether a found plaintext in Sanborn's papers counts as "solving" Kryptos depends partly on how much the artist cared about method versus result.

🔨Sanborn's Workshop

Kryptos was fabricated under real workshop constraints, with assistants cutting and assembling hundreds of copper letters by hand. That production history matters because it explains why some anomalies may be artistic and others may simply be frozen-in-place fabrication drift. Public-art cryptography is not the same as a clean typeset cipher sheet; the medium itself becomes part of the evidence.

📚References
  • CIA, Kryptos Sculpture legacy page, on the full installation and the 1,735-letter count.
  • David Stein, The Puzzle at CIA Headquarters: Cracking the Courtyard Crypto, Studies in Intelligence 43(1), 1999.
  • Elonka Dunin, essays and FOIA-related Kryptos documentation; also coverage of Cyrillic Projector and the wider Kryptos-solving community.
  • Friedrich L. Bauer, Hans Link, and Elonka Dunin / community discussions on the Hill-cipher conjecture as summarized in Cryptologia scholarship.
← Previous Voynich Manuscript