Skip to main content
AVOID.NET

Fact-check findings

What an automated fact-checker found when it re-read Payy Network — Noir/Barretenberg Verifier Flaw Post-Mortem against the sources the page cites. Only the most recent review is shown.

Read this first

These findings are produced by an automated reviewer, and its results vary between runs: the same page, checked three times on the same day, came back with 15%, 20% and 34% of its claims disputed, mostly because each run extracted a different number of claims. Treat what follows as leads, not rulings.

“Disputed” means the reviewer could not reconcile the claim with the evidence it cited. It does not mean the claim is false. “Unverifiable” means no reachable source settled it either way.

Nothing here changes the page on its own. A proposed correction is applied only after a human moderator approves it; until then the page reads as it did when reviewed.

disputed

3 claims

The reviewer could not reconcile the claim with the evidence it cited. This is a lead, not a ruling that the claim is false.

  1. #5[disputed][awaiting moderator]in the timeline
    “A GitHub security advisory and CVE (CVE-2026-48100) describing a soundness flaw in Payy's agg_agg aggregation circuit, patched in version 1.3.0, was published.”
    reviewerThe GHSA/CVE was published in September 2026 (2026-09-28)The primary advisory itself gives June 10, 2026, not September. The timeline's 2026-09 date and date_original 2026-09-28 (probably the archive date) conflict with it. The date field fix is in the companion finding.
    Proposed correction (not yet applied)
    A GitHub security advisory and CVE (CVE-2026-48100) describing a soundness flaw in Payy's agg_agg aggregation circuit, patched in version 1.3.0, was published on June 10, 2026, according to the GitHub advisory page.
  2. #6[disputed][awaiting moderator]in the timeline
    “2026-09”
    reviewerTimeline GHSA date field should be June 2026Date field conflicts with the primary source. The entry's position in the chronology would also need to move before the exploit.
    Proposed correction (not yet applied)
    2026-06
  3. #7[disputed][awaiting moderator]in section: Alternative/Overlapping Technical Account: agg_agg Circuit Flaw (CVE-2026-48100)
    “with at least one aggregator listing a June 2026 publication date that would predate the September 24, 2026 exploit. This date inconsistency could not be resolved from available sources and may reflect a GHSA 'introduced version' backdating convention rather than actual pre-exploit public disclosure”
    reviewerThe June 2026 date may reflect a backdating convention, not pre-exploit disclosureThe page treats the June date as an aggregator discrepancy and speculates a backdating convention. The official advisory shows June 10, 2026, so that speculation is unsupported. The proposed correction replaces only the disputed clause. Whether the advisory was public in June or the page date reflects a later publication of a draft is not established.
    Proposed correction (not yet applied)
    including the GitHub advisory page itself, which lists a June 10, 2026 publication date that would predate the September 24, 2026 exploit. The advisory text references a May 5, 2026 balance check and describes a proof of concept on a local test deployment

unverifiable

2 claims

No source the reviewer could reach confirms or contradicts the claim.

  1. #3[unverifiable][awaiting moderator]in section: Post-Mortem: Noir/Barretenberg Verifier Attribution
    “Secondary reporting citing the post-mortem states that Barretenberg's small-subgroup IPA verifier was missing a boundary-opening check (described as 'A(1) = 0'), that Aztec fixed this in a commit dated May 14, 2026 and shipped the fix in Barretenberg v5.0.0 on July 13, 2026”
    reviewerPost-mortem technical details: fix commit May 14, 2026, v5.0.0 shipped July 13, pinned version, no notificationThe cited SpendNode article does not contain the commit date, release date, version pin or notification claims. The page already flags low confidence, but the attribution to SpendNode appears wrong (the fetch may have been summarized, so partially a limit of the fetch).
  2. #4[unverifiable][awaiting moderator]in the timeline
    “Aztec reportedly fixed a missing boundary-opening check (A(1) = 0) in Barretenberg's small-subgroup IPA verifier, per Payy's post-mortem as summarized by a secondary source.”
    reviewerTimeline: Aztec fixed missing check on 2026-05-14 per SpendNodeCited source does not show the date.

stale

2 claims

The claim was accurate when written but events since have overtaken it.

  1. #10[stale][awaiting moderator]in section: Official Response and Denials
    “As of the most recent reporting reviewed, Payy had not announced a timeline for restoring service or confirmed whether affected users would be made whole.”
    reviewerPayy had not confirmed whether affected users would be made wholeSuperseded by reporting from Oct 13, 2026. A related sentence in sections[5] has the same issue.
    Proposed correction (not yet applied)
    Payy subsequently opened a 30-day claims window for affected users, with eligible claims to be paid in USDC on Ethereum, and said Payy Wallet would be fully shut down after the claims window closes.
  2. #11[stale][awaiting moderator]in section: Prior Security History
    “whether affected non-custodial deposits would be reimbursed, or a timeline for resuming service”
    reviewerPayy had not disclosed whether affected deposits would be reimbursed or a timeline for resuming serviceSame supersession as sections[1]. The replacement would need the sentence's preceding words adjusted by an editor.
    Proposed correction (not yet applied)
    or the number of other affected users. Payy has since opened a 30-day claims window and said Payy Wallet will be fully shut down after it closes

partially supported

2 claims

The cited evidence supports part of the claim but not all of it.

  1. #2[partially supported][awaiting moderator]in section: Fund Movement and Laundering
    “Multiple outlets reported that the stolen USDC was routed through the Railgun privacy protocol, converted into approximately 683 ETH”
    reviewerStolen USDC routed through Railgun into about 683 ETHThe 683 ETH figure is supported, but the Unchained article attributes the conversion to UniswapX, not Railgun. The Railgun claim was not verified.
  2. #12[partially supported][awaiting moderator]in section: User Impact and Unresolved Questions
    “asserting the company had patched a critical zero-knowledge circuit logic flaw in June 2026 before it was exploited”
    reviewerSecondary source says Payy patched a critical circuit flaw in June 2026 before the exploit (second security issue)The page says this could not be corroborated, but the GHSA on the same page is dated June 10, 2026 and rates a Critical circuit flaw, which corroborates a June disclosure. I did not verify CryptoBriefing's text.

confirmed

4 claims

The cited evidence supports the claim as written.

  1. #1[confirmed][no action needed]in section: The Exploit
    “A second, smaller transaction roughly five hours later (reported as occurring around 09:30 UTC) released a further 90,202.820016 USDC to the same address”
    reviewerDrain of roughly 1,828,589 USDC at 04:21 UTC and 90,202 USDC around 09:30 UTC to the same address, total about $1.92MTimes and amounts match Unchained. Block number 26044909 and the burn substitutor detail were not checked in the fetched article.
  2. #8[confirmed][no action needed]in section: Alternative/Overlapping Technical Account: agg_agg Circuit Flaw (CVE-2026-48100)
    “prior to version 1.3.0 the agg_agg circuit forwarded the compacted message stream from inner proofs into a public messages array without constraining the unused tail of that array to zero”
    reviewerAdvisory details: agg_agg does not constrain unused tail, forged burn, credit loopghost, patched v1.3.0, affected zk-circuits-v1.1.12Technical description matches the advisory. The advisory notes the current deployment limited submissions to an allowlisted prover, which the page's wording about a registered prover is consistent with.
  3. #9[confirmed][no action needed]in section: Official Response and Denials
    “The company subsequently stated the incident was 'NOT a compromised key, social engineering or exploit of our off-chain infrastructure,'”
    reviewerPayy stated it was not a compromised key, social engineering or off-chain exploit; notified law enforcement; engaged an audit firmMatches Unchained.
  4. #13[confirmed][no action needed]in the timeline
    “A second transaction several hours after the first drained a further approximately 90,202 USDC to the same receiving address, bringing the combined total to roughly $1.92 million.”
    reviewerTimeline: 2026-09 second transaction brings combined total to roughly $1.92MConsistent.
How this fits together. The reviewer reads the published page and its cited sources and records one finding per claim. A human moderator decides whether each proposed correction is applied; those decisions, and the score changes they cause, appear in the audit log. Earlier review runs are not shown here; only the latest reflects the page as it stands.