Verified Facts
The short list of statements on this site that were recomputed from public raw data and confirmed by the September 2026 re-check. Everything else on the site is either marked as a hypothesis or listed as an open question on the re-check page.
Verified Facts
A fact listed here is true as stated. Being true does not make it meaningful: several of these are exact arithmetic that earlier versions of the site over-interpreted. The right-hand column says what each fact does not show.
The Anna Matrix (as published)
| Fact | What it does not show |
|---|---|
The published file is a 128 x 128 table of signed bytes, compiled by hand from replies of the @anna_aigarth bot to x+y=?; the reply to x and y is in row 63 - y, column x + 64. All 1,218 numeric bot replies in our records (948 distinct queries) match this lookup | It is not a weight file, not source code and not a memory dump. The replies and the table come from the same collection, so the match is consistency, not independent proof |
16,316 of 16,384 cells satisfy M[i][j] + M[127-i][127-j] = -1 | The 68 exceptions are not a hidden layer; most are entry errors (chapter 36) |
| The matrix has full rank 128 | No meaning beyond linear algebra |
| The 26 largest Walsh coefficients reproduce about 81% of the signs; a 32-row kernel reproduces 94.5% | The table is mostly a shallow, almost linear function of the input bits |
For 768 tested ternary input vectors, iterating sign(M v) ends in two period-4 cycles that are negatives of each other | Not proven for all inputs. The seven positions where these cycles are not antipodal depend on the column 30 entry error |
| Three diagonal cells have absolute values 75, 101 and 121, which are the ASCII codes of "K", "e", "y" | Under a fair chance model this is not unusual |
| The trace of the table as typed is 137 | It comes from one cell pair that is very likely an entry error; the corrected table has trace -64 |
Bitcoin
| Fact | What it does not show |
|---|---|
| The Genesis block is 285 bytes; its nonce is 2083236893 | Nothing about Qubic |
| The Genesis output public key is a valid secp256k1 point | It is not an executable script and hides no second program |
Block 264 (13 January 2009, 00:19:40 UTC) pays to the public key whose address is 1CFBdvaiZgZPTZERqnezAtDQJuGHKoHSzg | Among 21,953 early keys, 17 addresses start with 1CF, while chance predicts 16.5. See chapter 17 |
The public key of block 4 gives an address starting with 15ubic | A prefix like that is expected somewhere among thousands of early addresses |
| 625,284 = 283 x 47² + 137, and 625,284 mod 16,384 points to row 21, column 4 | Exact arithmetic with numbers chosen after the fact |
Qubic and Aigarth
| Fact | What it does not show |
|---|---|
| Qubic derives a private key from a seed with two KangarooTwelve steps (seed, subseed, private key) | Nothing about Bitcoin keys |
The public Aigarth code contains a unit ITUClArithmeticAdditionIntI2x7O8 with exactly the bot's input and output format; such units map complemented inputs to complemented outputs | That the bot is exactly this unit. The table fits a small trained unit of this class, which is a supported hypothesis (chapter 36) |
| The Anna.exe versions 1 and 5 to 8 evolve ternary lookup networks on signed 7-bit addition | They are not the program behind the bot's replies |
| Anna 0.22.64 (September 2026) contains a table of 9,456 real hourly BTCUSDT directions, 2 June 2025 to 30 June 2026 | That Anna can predict prices (chapter 37) |
Come-from-Beyond
| Fact | What it does not show |
|---|---|
| CFB has written "I'm not Satoshi" several times, with message IDs in chapter 17 and Appendix D | A denial is not proof either way |
| CFB has named 13 April 2027 several times: "No ETA other than April 13, 2027" (1394734677935521824), "It's 5 years after launch of Qubic" (1250064529677750287), and "I can reveal this info on 13th of April 2027" followed by "And you'll see who is really behind CFB acc" (1426096720873521214, 1426096783318319164) | The messages do not say what exactly happens on that date. No event was announced for 3 March 2026 |