2023 · United States · J - Information and communication · DDoS attack causing platform outage
Analysed 2026-11-06 by Hermes
Open this case in the tool →DDoS attack causing platform outage
No DV evidenced
Alternative: none defensible
(1) The expectation at a platform of X's size and criticality is implicit rather than documented: as a globally available social platform and (in the dataset's terms) a digital provider it is assumed to keep serving users under attack. No primary source - no X statement, no regulator or CERT finding, no forensic report - documents what availability or DDoS posture X had or decided, because X never publicly acknowledged this incident and X Corp. was then a private company with no incident-disclosure duty. (2) What happened: on 29 August 2023 the hacktivist group Anonymous Sudan announced a DDoS on its Telegram channel at 01:31 and took X offline in more than a dozen countries for more than two hours, with nearly 20,000 outage reports logged in the US and UK; the BBC describes the technique as the same blunt and relatively unsophisticated flooding the group is known for; access returned within hours. (3) Why this stays Unknown: the reporting confirms impact only. The entry vector is known and is entirely the attacker's move - a volumetric DDoS - and reasoning from an attacker's technique, or from the two-hour outage alone, to a victim-side knowing-doing or doctrinal-material gap is the rejected form of inference; the method spec's decision rule requires a concrete, verifiable victim-side failure point on one of the axes, and none is documented at any lifecycle stage. Detection and restoration evidently worked (the outage lasted about two hours), egress was not in play (no data breach or exfiltration), recovery is not documented, and the response stage shows only silence, which for a private company under no notification duty is not itself a documented gap. A capacity or availability-protection reading would rest only on the coder's inference from the outage duration, so no DV-Type is defensible and no risk marker is coded. As a secondary observation the incident shows the group's documented capability to disrupt even very large platforms, but the DV is assessed on the victim side and X's side of the picture is undocumented. (The case also illustrates the framework's limit: roughly all that is verifiable here is the attacker's claim plus outage telemetry.)
Case-file data quality: the Victim column literally holds the single letter X, which matches the description and receiver_name (the platform X, X Corp.), so the row is internally consistent - X here is the company name, not a placeholder. The initiator field's "< Killnet" suffix reflects the 2023 false-flag theory (reported by the BBC and SOCRadar) that Anonymous Sudan is a Killnet front; the 2024 US indictment charges two Sudanese nationals and CrowdStrike assesses the group as independent but collaborating with Killnet - the Killnet-subordination label in the dataset is stronger than the evidence. The case file's other source URLs were not carried as citations: the t.me Telegram post is an attacker channel (policy: attacker claims cited only as reported by legitimate press), the SOCRadar page is primarily about the group's later Telegram attack and only mentions the X incident in passing, and the hackread page is about the November 2023 OpenAI/ChatGPT DDoS, a different incident. Disclosure was by the attacker; X Corp. was private in 2023 (Twitter Inc. merged into it in April 2023, TechCrunch reporting) and issued no public statement, so there is no primary victim source to research; no regulator or law-enforcement action names this specific X incident (the October 2024 DOJ indictment covers the group's campaign generally). Atypical note for the dataset: attacker-disclosed disruption case; Unknown here reflects undocumented victim-side mechanism, not an attacker-side target. Country from eurepoc_extra.receiver_country (United States, X Corp. headquarters), Year from start_date (29.08.2023).