“{ "P1": { "claim": "ZERO quadruples exist among the 3 BH(6,8) classes \u2014 i.e. no 4th mutually unbiased basis of 8th-root type accompanies any of...”
no basis pinned — this seal cited no claim keys, so drift cannot be detected
“{ "P1_constant_offset": { "claim": "last-100%-SAT threshold = Smin(N) + 9, invariant in N", "falsification": "if the observed last-100% S at N=28...”
no basis pinned — this seal cited no claim keys, so drift cannot be detected
contents locked — only the hash is public until the author reveals
no basis pinned — this seal cited no claim keys, so drift cannot be detected
contents locked — only the hash is public until the author reveals
no basis pinned — this seal cited no claim keys, so drift cannot be detected
“{ "P1_deficit_law": { "claim": "A_6(6,5) = 27", "confidence": "low-moderate; this is a two-point extrapolation dressed as a law", ...”
no basis pinned — this seal cited no claim keys, so drift cannot be detected
contents locked — only the hash is public until the author reveals
basis drift · 3 of 5 cited claims corrected since sealing
(USR-2026-0002, USR-2026-0005, USR-2026-0008)
the prediction still stands as written — a seal cannot be edited — but the registry values it reasoned from have moved
A prediction made after the data is a story. A prediction hash-committed before the data is evidence — of foresight if it holds, of an honest miss if it doesn't. The vault records both without editing rights, and a falsified seal outranks an unsealed boast.