2021 · Indonesia · Q - Human health and social work activities · Mass data theft sold on a hacker forum
Analysed 2026-11-07 by Hermes
Open this case in the tool →Mass data theft sold on a hacker forum
Strategy to Leadership
Evidence coverage: 3/6 stages, weight 20/60
Decisive stage: response (tie-break: post-breach stage) · Alternative: Strategy to Execution
(1) BPJS Kesehatan holds the national health-insurance register of virtually every Indonesian and led with a doctrinal security posture: an ISO 27001-certified security management system, a Security Operation Center it said ran 24 hours a day, and cooperation with BSSN, Kominfo and the police; the statutory framework (PP 71/2019, Menkominfo Reg 20/2016) required operators to report personal-data protection failures at the first opportunity and notify data owners in writing. (2) In practice the theft surfaced from outside: the account 'Kotz' posted and sold the dataset on RaidForums on 12 May 2021; BPJS's claimed detection apparatus noticed nothing (Reuters: the expert noted a leak that 'largely went undetected'), validity was established by independent experts checking samples against BPJS's own public verification site and by Kominfo's own analysis, and BPJS leadership spent the following months publicly denying that the data could be its own while it was still being sold. The entry vector was never disclosed and is not load-bearing. (3) The expectation that was not acted on sat at the strategic/leadership level - the statutory duty to report and notify and the compliance posture leadership itself communicated - and the chain broke at leadership, not in execution: no accurate communication was ever decided, the president director instead communicated a public account contradicted by the regulator's finding and independent verification, and the case file of public statements shows the doctrinal column turned against itself (the secondary pattern is a Strategy to Execution reading - certified ISMS and claimed SOC never realised in operations - which carries the largest share; the classification rests on the response stage under the post-breach tie-break). (4) RM39 (Strategy to Leadership lens): leadership disclosure conduct diverged from the legal expectation - a press-conference denial of attribution instead of the required report and written notice to data owners, with the public disclosure of exposure in effect forced by the seller's forum post (causal role: Constraining). RM44 (Strategy to Leadership lens): leadership presented a specific posture - ISO 27001 certification, a 24/7 SOC, layered security - that the documented reality contradicts, since neither the intrusion nor the wholesale export of its register was detected by BPJS (causal role: Predisposing). RM34 (Strategy to Leadership lens): detection readiness that leadership reported as standing did not function - time-to-detect was effectively zero for the victim and infinite for the agency, which learned of the leak from the market, so oversight produced no meaningful detection capability (causal role: Obscuring). RM16 (Strategy to Leadership lens): leadership commitments on data responsibility were not operationalised - salary histories and profile photos for the whole population, including deceased citizens, held centrally and exfiltratable at once, while consultants described requirements being met for audit purposes only and never upgraded afterwards (causal role: Amplifying).
RM34 — Monitoring / detection / alert-handling gapRM16 — Data governance / minimisation gapRM39 — Breach notification / disclosure failureRM44 — Public security-claim discrepancyinformationisbeautiful row: the one-line description matches BPJS Kesehatan (ID numbers, salary, phone numbers of the whole population); the row's source_url (KrASIA) names the agency, so the victim as written ('Indonesia's health agency') is confirmed and refined. Scale caveat: the 279M figure is the seller's claim; Kominfo verified only a 100,002-record sample as strongly suspected to be identical to BPJS data, and BPJS never acknowledged the breach even in later reporting - the row's 'every single man, woman and child' rests on the attacker's claim as reported by mainstream press. No official investigation outcome was ever published (Kominfo's formal decision was still pending on 25 Oct 2021), so the classification rests on the documented response conduct, not on a regulator finding. Actor attribution: none - police were unable to identify the perpetrators; an unattributed actor does not lower confidence. Security policy: the RaidForums posts and the leaked dataset were not accessed; attacker claims are cited only as reported by established press. Data-quality note on the case file's source_url: the URL as stored carries a trailing ' /' and had to be cleaned to open. Related rows: the case file carries none and no earlier result for this victim exists in the output directory. Link check: reuters.com blocks automated access (HTTP 401); the page was opened and read during research. Link check: en.tempo.co blocks automated access (HTTP 403); the page was opened and read during research.