{
 "public_id": "AIR-2026-0203",
 "incident_id": "2026-08-straiker-pod9-sre-agent-ransomware",
 "date_disclosed": "2026-08-24",
 "date_occurred": null,
 "title": "Escape from Pod 9: poisoned telemetry makes an AI SRE agent deploy a privileged ransomware container on EKS",
 "class": "research_demo",
 "attack_surface": "autonomous_ops",
 "initial_vector": "indirect_prompt_injection_telemetry",
 "agent_capability_abused": [
  "kubernetes_api_access",
  "container_deployment",
  "telemetry_ingestion"
 ],
 "target_product": "AI SRE agent built on Amazon Bedrock monitoring an EKS workload via OpenTelemetry",
 "target_vendor": "Amazon Web Services (Bedrock) / custom agent",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "ransomware_deployment",
  "privilege_escalation",
  "cluster_compromise"
 ],
 "blast_radius": "Kubernetes clusters where an AI SRE agent holds deployment permissions and ingests untrusted telemetry",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "telemetry_poisoning",
   "kubernetes",
   "privileged_container",
   "autonomous_ops"
  ]
 },
 "defense_present": null,
 "defense_bypassed": true,
 "fix_shipped": null,
 "primary_sources": [
  "https://www.straiker.ai/blog/ai-sre-agent-prompt-injection-kubernetes-ransomware"
 ],
 "verified_quote": "No software vulnerability was exploited. No credentials were stolen. No zero-day was used.",
 "confidence": "high",
 "notes": "Author Eduardo Arrucha (Straiker STAR Labs). Injection carried in failed-login records flowing through OpenTelemetry into the agent's context. Straiker is explicit that the flaw is agent design, not Bedrock or the model.",
 "risk_category": "security_fraud",
 "risk_secondary": [],
 "risk_subcategory": "Prompt Injection",
 "risk_sub_secondary": [
  "Unauthorized Access & Account Takeover",
  "Ransomware & Extortion"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "the SRE agent deployed the privileged ransomware container",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "indirect_prompt_injection_telemetry",
 "guardrail_kind": [
  "unknown"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "unknown",
 "model_provider": "unknown",
 "model_name": "",
 "detection_mechanism": "external_researcher",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "privilege_escalation",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [
  "confused_deputy"
 ],
 "eai_reason": "impact rule: privilege_escalation",
 "asi_curated": [
  "ASI01"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI01 | added ASI03 (signal=identity)",
 "submitted_by": "EAI Internal"
}