What was already known
The standards specified the public seeds and the hash-based procedure deriving curve parameters from them, while the origins of the seeds themselves remained unexplained. Earlier public searches and the 2023 bounty largely pursued a remembered passphrase-hashing story. Truncated linear-congruential-generator state recovery and LLL are established techniques; the reported contribution is identifying this particular historical construction.
AI involvement
Anthropic says an employee asked an internal research model to investigate without supplying the earlier passphrase clue. The account credits Claude with identifying generator structure, recovering truncated states and testing candidate recipes; staff pursued and validated the findings. Prior staff searches were unsuccessful. The public account supports this attribution but does not establish a fully audited, human-free discovery history.
The institutional account attributes the generator identification, truncated-state recovery and retained recipes to Claude; employee questions and validation do not document human-free work.
Importance
Provides a compact, reproducible computational explanation for a long-standing parameter-provenance question in widely used elliptic-curve cryptography.
Limitations
Matching public constants does not uniquely establish historical intent, reveal discarded timestamp searches or explain why some viable curve candidates were skipped. It is not a cryptographic break, an attack on ECDSA or a proof of curve security or absence of backdoors. B-163 and several base-point origins remain unresolved. Six additional base-point reconstructions are author-reported and were not reproduced in this review. AIProofWatch ran the seed arithmetic: all fourteen results match the public bounty list, and nine NIST results also match the original FIPS 186-2 tables; the five non-NIST ANSI values were not checked against an independently retrieved ANSI standard. The institutional account does not provide a complete interaction log or an independent human assessment of every provenance claim.
Technical verification
Only the stated computational, numerical, conditional or component checks are documented for this scope; the full claimed endpoint has not received a documented complete rigorous check in the reviewed evidence. Matching public constants does not uniquely establish historical intent, reveal discarded timestamp searches or explain why some viable curve candidates were skipped. It is not a cryptographic break, an attack on ECDSA or a proof of curve security or absence of backdoors. B-163 and several base-point origins remain unresolved. Six additional base-point reconstructions are author-reported and were not reproduced in this review. AIProofWatch ran the seed arithmetic: all fourteen results match the public bounty list, and nine NIST results also match the original FIPS 186-2 tables; the five non-NIST ANSI values were not checked against an independently retrieved ANSI standard. The institutional account does not provide a complete interaction log or an independent human assessment of every provenance claim.
Human understanding
Named human researchers explain or reconstruct the scoped central mechanism in the cited paper/account; credit for that explanation is separate from technical checking. Anthropic says an employee asked an internal research model to investigate without supplying the earlier passphrase clue. The account credits Claude with identifying generator structure, recovering truncated states and testing candidate recipes; staff pursued and validated the findings. Prior staff searches were unsuccessful. The public account supports this attribution but does not establish a fully audited, human-free discovery history.
Correction and publication history
Primary institutional statement, public computational artifact, original NIST tables, earlier search context, human/model roles and limitations reviewed on 17 October against every current catalog/update page. Genuinely new scoped computational result; no formal proof, outside validation or cryptographic security claim.
Read this published revision