Seal fingerprints every AI model and dataset, then records its lineage on-chain — so anyone can audit provenance before they deploy.
every entry is hashed, timestamped, and written to the chain. no more trusting a model card — verify it.
| model / dataset | seal | fingerprint | license | status | sealed at |
|---|---|---|---|---|---|
| whisper-seal-v2speech-to-text | e9a2c4…7d1f | mit | verified | 2 min ago | |
| llama3-8b-instructchat | 7c41f0…b2e8 | llama2 | auditing | 9 min ago | |
| stable-diffusion-xlimage | a3b9d1…c0e4 | openrail | verified | 17 min ago | |
| gpt-neox-20btext | 5d8e2f…a1b7 | apache-2.0 | verified | 31 min ago | |
| common-crawl-cleandataset | b7f3a9…0c2d | cc-by | new | 1 hr ago | |
| fineweb-edu-v3dataset | 2c8a7e…f4b6 | odc-by | verified | 3 hr ago |
every checkpoint links back to the data it was trained on — and forward to where it is deployed.
your model or dataset is hashed with sha256 — a fixed fingerprint that changes if anything changes.
we link the fingerprint to its parents: training data, base model, license, contributor.
independent validators check the fingerprint and sign it — cryptographically.
the attestation is written on-chain. it is now public, permanent, and auditable.
verified (audited + signed) · auditing (in review) · new (awaiting first attestation). deploy teams can filter by state.
mit, apache-2.0, llama2, openrail, cc-by, odc-by — every seal carries its license, machine-readable.
verified datasets and clean fingerprints earn $SEAL.
every audit is published with its signature — the full history, not a summary.
validators stake $SEAL to attest fingerprints — slashed if they sign a bad one.
9,412 validators across 74 regions sign every seal. no single party controls provenance.