CSAIS
Hoax warning· OJK· TEST DATAOnline scamIndonesia · Indonesian

Kemkomdigi Ungkap 66 Persen Masyarakat Indonesia Pernah Jadi Korban Online Scam

First seen 28 Sept, 18:17 WIBLatest 30 Sept3 articles · 3 publishers
victim: persen masyarakat indonesia pernahVerify evidence
29
Confidence index · verifiedcomputed by the pipeline (V0.7)low
Verified by an institution decision (D4)machine score 0.74 → final 0.29 (verification weight 60%)
Corroborated by distinct domains
Independent sources
Victim identified
Full text available
Clustering consistency
Status and preventive steps
Hoax warningOfficial decision by OJKTEST DATA

Institution's reason: Informasi tidak benar (hoaks). DATA UJI: keputusan contoh untuk pengujian alur, bukan pernyataan resmi OJK.

Preventive steps
  • Do not forward this information; the official institution has stated it is false.
  • Check similar news on the institution's official site and accounts before sharing.
Literacy audienceBank customers and e-wallet usersJob seekers and fresh graduatesSocial media and messaging users
Literacy topics
  1. Checking information before sharing it
  2. How to report and keep evidence

Report toAduan Konten Komdigi

Extracted claims

machine value (D1); public and institution reviews below
Victimconfidence 0.90
Persen Masyarakat Indonesia Pernah
  • D2 Public surveymatches
Actornot found
not identified
Attack typeconfidence 0.90
Online scam
  • D2 Public surveymatches
Event dateconfidence 0.50
2026-09-30
LocationIndonesia
  • D2 Public surveymatches
Target groupIndividuals
Status
  • D1 Machinehigh trust
  • D2 Public surveymatches
  • D3 Institutionrefuted
  • D4 Official recordrefuted
Conclusion
  • D1 MachineHappened
  • D2 Public surveyNot yet certain
  • D3 InstitutionHoax
  • D4 Official recordHoax

Source chronology

3 articles, by publication time

28 September 20262 articles

  1. 18:17 WIBFirst report
    Kemkomdigi Ungkap 66 Persen Masyarakat Indonesia Pernah Jadi Korban Online Scamtopik.idPersen Masyarakat Indonesia Pernah
  2. 19:25 WIB+1 hfull text hashed
    Komdigi Ungkap 66 Persen Masyarakat Indonesia Pernah Jadi Korban Penipuan Digitalntvnews.idPersen Masyarakat Indonesia Pernah

30 September 20261 articles

  1. 07:05 WIB+37 hfull text hashed
    Komdigi: 66 Persen Masyarakat Indonesia Pernah Jadi Korban Penipuan Digitalmediaindonesia.comPersen Masyarakat Indonesia Pernah
Relations between articles
1 pairs likely independent

Evidence

Hashes prepared2full text hashed 2
first evidence_uid1e972cce30743449ec31e217
2 of 2 evidence records are in Merkle batches

Flow comparison D1–D4

each flow's preventive output; this is where they differ · D1 changed 3 times · since 29 Sept
D1Machine12 OctD2Public survey12 OctTEST DATAHelp reviewD3Institution1 OctTEST DATAD4Official record1 OctTEST DATA
Literacy audienceBank customers and e-wallet usersJob seekers and fresh graduatesSocial media and messaging usersBank customers and e-wallet usersJob seekers and fresh graduatesSocial media and messaging users––
Literacy topics
  1. 1.Safe online transactions
  2. 2.Social engineering and phone scams
  3. 3.How to report and keep evidence
  1. 1.Safe online transactions
  2. 2.Social engineering and phone scams
  3. 3.How to report and keep evidence
––
Preventive steps
  • Turn on two-step verification for email, banking, and social media.
  • Verify the other party through official channels before transferring money.
  • Check the account or phone number on cekrekening.id before paying.
Report to Cek Rekening, Patroli Siber Polri
  • Turn on two-step verification for email, banking, and social media.
  • Verify the other party through official channels before transferring money.
  • Check the account or phone number on cekrekening.id before paying.
Report to Cek Rekening, Patroli Siber Polri
––
Institution's reason––Informasi tidak benar (hoaks). DATA UJI: keputusan contoh untuk pengujian alur, bukan pernyataan resmi OJK.Informasi tidak benar (hoaks). DATA UJI: keputusan contoh untuk pengujian alur, bukan pernyataan resmi OJK.Verify this record in the ledger

Summary across incidents

Related incidents

semantic similarity of the coverage, across languages