Skip to main content
AVOID.NET
← Radiant Capitalreviewed 2026-10-27 · 10 claims checked

Fact-check findings

What an automated fact-checker found when it re-read Radiant Capital 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.

unverifiable

3 claims

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

  1. #4[unverifiable][awaiting moderator]in section: October 2024 Multisig Compromise and North Korea Attribution ($50M)
    “The attack originated on September 11, 2024, when a Radiant developer received a Telegram message impersonating a known former contractor, purportedly sharing a PDF about smart contract auditing.”
    reviewerAttack began Sept 11, 2024 with Telegram message to a developerSpecific day not found in the sources I could fetch; the Medium incident update was not checked. Month is supported.
  2. #6[unverifiable][awaiting moderator]in section: Multisig Security Model Failure
    “After the hack, Radiant's DAO raised the multisig threshold to 4-of-7 (57%), reducing both the target pool and increasing the share of devices an attacker must compromise.”
    reviewerDAO raised multisig threshold to 4-of-7 after the hackNot found in the cited Halborn post; other cited sources not checked.
  3. #10[unverifiable][awaiting moderator]in section: Team Identity and Transparency
    “Radiant Capital was founded in 2022 and is reportedly headquartered in Hong Kong.”
    reviewerReportedly headquartered in Hong Kong; founders undisclosedNansen and radiant.capital not fetched; the page hedges with 'reportedly'.

stale

1 claim

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

  1. #8[stale][awaiting moderator]in section: Protocol Recovery and Current Status
    “Radiant also began migrating its token to the OFT (Omnichain Fungible Token) standard as part of a V2 relaunch. The protocol's official website remains active at radiant.capital, but community sentiment remains predominantly bearish given the severity of losses and the absence of substantive fund recovery.”
    reviewerPage presents the protocol as active with a V2 relaunch, omitting its wind-downMore recent reporting supersedes the V2 relaunch framing.
    Proposed correction (not yet applied)
    Radiant also began migrating its token to the OFT (Omnichain Fungible Token) standard as part of a V2 relaunch. In June 2026 the Radiant DAO announced a wind-down, stating it had no viable path forward after failing to recover the stolen funds; the protocol moved to a maintenance state in which the front end and smart contracts remain accessible for withdrawals and repayments. Community sentiment remains predominantly bearish given the severity of losses and the absence of substantive fund recovery.

partially supported

2 claims

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

  1. #2[partially supported][awaiting moderator]in the timeline
    “2024-12-09”
    reviewerMandiant formally attributed the attack on 2024-12-09Attribution was disclosed in Radiant's Dec 6 update (already a separate timeline entry); Dec 9 is a CoinDesk publication date, not a Mandiant action date. The entry duplicates the Dec 6 one.
  2. #3[partially supported][awaiting moderator]in section: January 2024 Flash Loan Exploit ($4.5M)
    “On January 2, 2024, Radiant Capital's newly activated native USDC lending market on Arbitrum was exploited for approximately $4.5 million (roughly 1,900 ETH) within six seconds of the market going live.”
    reviewerFlash loan exploit on Jan 2, 2024 of ~$4.5M (about 1,900 ETH) within six secondsDate, amount and ETH figure supported; 'six seconds' not found in the Cointelegraph article. Note the section heading calls it a 'flash loan' exploit while the body describes a rounding-error exploit.

confirmed

4 claims

The cited evidence supports the claim as written.

  1. #1[confirmed][no action needed]in section: October 2024 Multisig Compromise and North Korea Attribution ($50M)
    “Cybersecurity firm Mandiant, retained by Radiant for forensic investigation, attributed the attack with high confidence to UNC4736, also tracked as Citrine Sleet or AppleJeus, a threat actor assessed to have a DPRK nexus with ties to North Korea's Reconnaissance General Bureau (RGB).”
    reviewerMandiant attributed the October 2024 attack to UNC4736 (Citrine Sleet/AppleJeus), DPRK nexus, RGBSupported by SecurityWeek and CoinDesk.
  2. #5[confirmed][no action needed]in section: Multisig Security Model Failure
    “Radiant Capital operated a 3-of-11 multisig scheme at the time of the October 2024 attack, meaning only three of eleven possible signatories were required to authorize critical protocol transactions.”
    reviewerRadiant operated a 3-of-11 multisigSupported by Halborn.
  3. #7[confirmed][no action needed]in section: Post-Hack Fund Recovery and Laundering
    “On October 23, 2025, blockchain analytics firm CertiK reported that the exploiter deposited 2,834.6 ETH (approximately $10.8 million) into the Tornado Cash mixer.”
    reviewerExploiter swapped ~3,091 ETH for 13.26M DAI in Aug 2025 and CertiK reported 2,834.6 ETH (~$10.8M) into Tornado Cash on Oct 23, 2025Confirmed. The PeckShield Oct 31 claim was not independently fetched.
  4. #9[confirmed][no action needed]in section: Post-Hack Fund Recovery and Laundering
    “As of late 2025, the majority of stolen funds had not been recovered.”
    reviewerMajority of stolen funds not recovered as of late 2025Consistent with the June 2026 wind-down reporting.
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.