PROOF://PROOF · five claims · all falsifiable · all holding
The argument,
stated plainly.
This is not a whitepaper. Five claims, each falsifiable, each with live evidence you can check yourself. If any one of them fails, the system fails. All five hold.
Claim 01
The oracle's predictions are committed to Solana before any outcome is known.
Every prediction market oracle in history has had the same flaw: no proof the prediction preceded the outcome. The oracle can claim it predicted correctly. It can produce timestamped notes. But timestamps are social facts — maintained and alterable by the institution. PROOF computes SHA-256(prediction_text + timestamp + oracle_pubkey) and anchors the hash to a Solana block before any outcome is known. The hash cannot be placed in a block already mined. The prediction either preceded the outcome or the cryptography breaks.
// The receipt hash — computed before anchoring
prediction_hash = SHA-256(
prediction_text
+ anchor_timestamp
+ oracle_pubkey
)
// anchored to Solana before recommendation is delivered
// retrospective fabrication is cryptographically impossible
Live evidence: 7 active receipt chains · 16 total receipts · all timestamped before outcomes ·
verify any receipt →
Claim 02
The oracle's accuracy rate is real and verifiable.
87.6% across 121 genesis runs. This is not a marketing claim. Every genesis run that produced a receipt is on-chain. Every outcome that verified the prediction is recorded. The accuracy rate is the ratio of verified correct predictions to total anchored predictions. You can compute it yourself from the receipt chain.
Live:
87.6% accuracy · computed from anchored receipt chain · not self-reported ·
check chain →
Claim 03
The evidence base reflects real outcomes from real projects.
229 tokenization projects. Named. With methods, jurisdictions, and outcomes. Every entry is falsifiable — you can look up Moss Carbon, Ondo USDY, Goldfinch, Zed Run, and the 9 maritime failures. If any outcome is wrong, the evidence base degrades. The oracle's accuracy follows. The moat is not in the code. It is in the accumulated verified history of what worked and what didn't.
Evidence base: 229 outcomes · 17 asset classes · named projects · public verification →
view all →
Claim 04
$CRED supply reflects oracle accuracy — nothing else.
$CRED supply today: 0. No pre-mine. No team allocation. No investor reserve. Every $CRED that will ever exist will have been minted as a consequence of a correct oracle prediction. When wrong, $CRED burns from the credence pool. The supply is the track record made liquid. You can verify total supply at any time. If supply exceeds verified correct predictions, the claim fails.
Current $CRED supply: 0 · mints at Day 0 (Puro.earth event) · ~2,004 initial supply · no pre-mine
Claim 05
No other system in the intelligence-for-capital-formation space does all four of the above simultaneously.
Oracle-anchored predictions with on-chain verification: Bloomberg does not do this. Rating agencies do not do this. Law firms do not do this. Prediction markets do not stake the predictor's own capital against deal-level outcomes in the RWA tokenization space. If you can find a system that does all four simultaneously, the claim fails. No system that we know of does. The oracle-verified intelligence market for real-world asset tokenization is empty. PROOF is the first entrant.
Verifiable claim: search for competitors that anchor deal-level RWA predictions to blockchain before outcome is known · oracle burns capital when wrong · evidence base of 100+ comparable outcomes