# LLMjacking via stolen cloud credentials (Sysdig)

- ID: llmjacking-sysdig-2024-05
- Reported: 2024-05-06
- Researcher: Alessandro Brucato / Sysdig TRT
- Organization: Sysdig
- AI role: ai_credential_theft (AI credential theft)
- Confidence in AI involvement: not assessed
- Research / borderline: false
- Last reviewed: 2026-11-05
- Entry page: https://openattacks.ai/attacks/llmjacking-sysdig-2024-05/
- Entry JSON: https://openattacks.ai/api/attacks/llmjacking-sysdig-2024-05.json
- Browse title: LLMjacking via stolen cloud credentials (Sysdig)
- Editorial summary: Attackers used stolen cloud credentials to probe ten AI services and resell LLM access through proxies, leaving victims to pay the inference bills.
- Evidence class (catalog annotation): Reported campaign
- Source-stated target environments: Cloud & SaaS
- Stage hunts and conditional defenses: https://openattacks.ai/api/defenses/llmjacking-sysdig-2024-05.json

## Technique
After a Laravel CVE-2021-3129 foothold, attackers stole cloud credentials and probed ten AI services (AI21, Anthropic, AWS Bedrock, Azure, ElevenLabs, MakerSuite, Mistral, OpenAI, OpenRouter, and GCP Vertex) to resell LLM access via reverse proxies while victims paid inference bills. Later Sysdig reporting showed 85,000+ Bedrock requests with Converse API abuse within 30 days of launch. Pillar Security separately reported the Bizarre Bazaar campaign targeting AI startups for unauthorized inference resale.

## Stack relevance (editorial)
Assessment: mapped. Editorial applicability reviewed against the catalog source account; not a scan or independent incident verification.
First action: Inventory AI service keys, scope permissions, rotate exposed tokens and investigate anomalous usage.
- Relationship: credential; tools: ai-api-keys; OS scope: not OS-specific / not stated; requires existing MCP connections: false.
  Relevance: The reported path uses stolen AI-service or cloud credentials to resell model access.
  Condition to check: Confirm the affected service and credential type in the source; consumer chat use alone is not the reported condition.
  Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking
A product mention alone is not a victim-side applicability mapping. Selection does not verify affected versions, permissions or compromise.

## Initial access
Vulnerable Laravel (CVE-2021-3129) leading to cloud credential theft

## AI tooling and models
- [AWS Bedrock](https://aws.amazon.com/bedrock/) [Claude](https://claude.com/product/overview) v2/v3
- OpenAI
- Anthropic
- Azure
- [Vertex AI](https://cloud.google.com/vertex-ai)
- others in probe list

## Other tooling
- LLM reverse proxy for resale

## Victims and sectors
unknown

## Target geography
- unknown

## Attribution
unknown

## Scale and metrics
Potential \>$46,000/day Claude 2.x costs cited; July 2024 spike 61k Bedrock requests in 3 hours

## Attack flow
Catalog reconstruction of the supplied source account. Unknowns and research caveats are preserved.
### step-1: Vulnerable Laravel provides a foothold [access]
The report identifies CVE-2021-3129 before credential theft.
Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking

### step-2: Cloud credentials are stolen [execution]
Attackers obtain keys and probe multiple AI providers.
Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking

### step-3: Inference access is proxied [ai]
Authorized model access is resold through attacker reverse proxies.
Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking

### step-4: Victim accounts incur inference usage [impact]
The report describes high-volume requests and potentially substantial costs.
Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking

Connections:
- step-1 -> step-2
- step-2 -> step-3
- step-3 -> step-4

## Stage defense and hunting recommendations
Editorial recommendations, not validated coverage. Source-controlled query starters are drafts. Custom-schema examples require telemetry collection and field mapping; absence of a mapping means not assessed. MagicSword policy opportunities require managed Windows scope and enforcement.
### step-1: Linux web worker → shell / interpreter
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: native
- SIGMA query download: /api/detections/linux-webshell.sigma.yml
- KQL query download: /api/detections/linux-webshell.kql
- SPL query download: /api/detections/linux-webshell.spl
- Hunt hypothesis: A PHP or web-server worker launches a shell, interpreter, or reconnaissance utility outside its normal baseline.
- Required telemetry: Linux process creation (EDR / auditd normalized to process fields); PHP / web access logs; File changes to web roots and PHP configuration
- Investigate: Trace the worker's children and correlate them with inbound requests.
- Investigate: Review auto\_prepend\_file, .bashrc changes, /dev/shm execution, and outbound tunnels where the report describes them.
- Limitation: Legitimate CGI, maintenance, and deployment scripts can match. The query is a process-tree starter, not a complete web-shell detector.
- No direct MagicSword prevention claim is recorded for this stage.

### step-2: Sensitive store access → credential use
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: custom_schema
- SIGMA query download: /api/detections/credentials.sigma.yml
- KQL query download: /api/detections/credentials.kql
- SPL query download: /api/detections/credentials.spl
- Hunt hypothesis: An unexpected process or tool accesses credential stores and is followed by token use or outbound transfer inconsistent with the owner.
- Required telemetry: File-read / API access telemetry for relevant secret stores; Process identity and command lines; Identity, token and outbound-network audit logs
- Investigate: Identify the actor, process and exact store touched, then pivot to token use from new devices or infrastructure.
- Investigate: Correlate reads with egress and rotate affected credentials after confirming exposure.
- Limitation: Defender DeviceFileEvents is not a general file-read audit trail. Collect suitable object-access, EDR or application logs; legitimate developer tools also read secrets.
- No direct MagicSword prevention claim is recorded for this stage.

### step-3: Bedrock discovery / key testing
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: native
- SIGMA query download: /api/detections/bedrock.sigma.yml
- KQL query download: /api/detections/bedrock.kql
- SPL query download: /api/detections/bedrock.spl
- Hunt hypothesis: A key enumerates Bedrock models from an unfamiliar origin, followed by unexpected model access or usage.
- Required telemetry: AWS CloudTrail management events; Bedrock invocation logging / relevant data events when explicitly enabled; Key inventory and inference usage
- Investigate: Group ListFoundationModels / GetFoundationModel calls by key, source IP and user agent; compare against normal application identities.
- Investigate: Correlate with invocation logs and cost. Discovery alone does not establish a stolen key or successful inference.
- Limitation: The starter queries cover discovery only. Do not assume default CloudTrail management logs contain InvokeModel / Converse requests or prompt content.
- No direct MagicSword prevention claim is recorded for this stage.

### step-4: Unexpected cloud / inference identity use
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: custom_schema
- SIGMA query download: /api/detections/cloud.sigma.yml
- KQL query download: /api/detections/cloud.kql
- SPL query download: /api/detections/cloud.spl
- Hunt hypothesis: A service key or role is used from an unfamiliar origin, gains broader access or causes inference usage outside its established baseline.
- Required telemetry: Cloud control-plane and identity audit; AI-service invocation / diagnostic logs when enabled; Usage, quotas, cost and key inventory
- Investigate: Pivot on key ID / principal, source network and user agent, then inspect role changes, token minting and service pivots.
- Investigate: Join inference usage and billing anomalies to the same principal; verify whether the behavior was authorized.
- Limitation: Cloud control-plane logs alone may not include model prompts or inference calls. No endpoint policy is claimed to stop cloud-only key misuse.
- No direct MagicSword prevention claim is recorded for this stage.


## MITRE ATLAS
Mappings are catalog annotations based on described behavior, not assignments or endorsements by MITRE.
- AML.T0012: Valid Accounts — Stolen cloud credentials are used to access inference services.
  Technique: https://atlas.mitre.org/techniques/AML.T0012
  Evidence: https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack ; https://www.sysdig.com/blog/growing-dangers-of-llmjacking

## Defensive notes
Monitor Bedrock ListFoundationModels+Invoke/Converse from long-term AKIA keys; rotate exposed keys; SCP restrict Bedrock

## Sources
- https://www.sysdig.com/blog/llmjacking-stolen-cloud-credentials-used-in-new-ai-attack — Sysdig: LLMjacking - Stolen Cloud Credentials Used in New AI Attack
- https://www.sysdig.com/blog/growing-dangers-of-llmjacking — Sysdig: Growing Dangers of LLMjacking
- https://www.pillar.io/blog/bizarre-bazaar-llmjacking-jan-2026 — Pillar Security: Bizarre Bazaar - LLMjacking Campaign Targeting AI Startups

## Source notes
2026 updates expand LLMjacking scope: Bedrock Marketplace abuse (FortiGuard), AI startup targeting (Pillar Bizarre Bazaar), and session token theft (Okta)
