Threat Actors Use LLMJacking to Monetize Compromised Cloud Models
Original title: The Rise of LLMJacking: How Threat Actors Monetize Compromised Cloud and Model Resources
Mandiant documented LLMJacking intrusions where threat actors hijacked enterprise LLM access for fraudulent charges exceeding $1M cumulatively across multiple sectors.
In 2026, threat actors hijacked enterprise access to LLMs via exposed endpoints or stolen cloud identity tokens including IAM credentials and session tokens. This led to large spikes in unauthorized API usage and fraudulent charges up to $100,000 in a day at impacted organizations in financial services, retail, technology, and transportation sectors across North America, South America, and Asia Pacific. Actors operated in specialized clusters with some using session hijacking to bypass MFA.
Unconfirmed: Specific threat actor cluster names and full attribution remain partially unconfirmed
- Fraudulent charges exceeded $100,000 USD in a single day at some organizations
- Cumulative charges reached $1 million USD before access revocation
- Actors harvested exposed LLM API keys from public GitHub repositories
- Threat actors enumerated cloud services and code repositories for additional credentials
- Two distinct operational clusters coordinated compromises and monetization via LLM Mixers
Impact: Monetization of compromised LLM resources via unauthorized API charges
| Product | Affected |
|---|---|
| LLM services and APIs | — |
- 1.Review and revoke exposed LLM API keys
- 2.Monitor for anomalous LLM API usage
- 3.Audit IAM credentials and Service Account keys
- Vendors
- Threat actors
- UNC6993
- Organizations
- financial servicesretailtechnologytransportation
- AI models
- Large Language Models
Indicators of compromise
auto-extracted · verify before use · export allAll| Type | Indicator | Context |
|---|---|---|
| domain | newapi.pro | ly siphon high-value commercial LLM access into the illicit newapi[.]pro resale ecosystem, directly draining the victim's corporat |
From the original article
12 min read at cloud.google.com
Written by: Christopher Romano, Codi Starks, John Wolfram, Josh Murchie, Mathew Potaczek, Garrett Mahoney, Mike Lombardi, K Singh, Rik Dolfing Introduction Mandiant responded to multiple incidents in 2026 in which threat actors hijacked enterprise access to Large Language Models (LLMs), a tactic known as "LLMJacking."
By Google Threat Intelligence Group · cloud.google.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.