● Covered by 2 sources · 2 reportsMedium impact2 neutral
Google Warns of LLMjacking Attacks Causing Up to $1 Million in Cloud AI Charges
Key points
- LLMjacking involves hijacking enterprise LLM access for fraudulent resource consumption.
- Attacks exploit stolen API keys, misconfigured environments, or exposed endpoints.
- Fraudulent charges can exceed $100,000 daily, with cumulative costs up to $1 million.
- Attackers sell unauthorized LLM access through "LLM mixers."
- Impacted entities are in North America, South America, and Asia Pacific across various sectors.
LLMjacking Attacks Lead to Significant Cloud Costs
Google's Threat Intelligence Group (GTIG) and Mandiant have reported a new type of attack, dubbed "LLMjacking," where threat actors gain unauthorized access to enterprise Large Language Models (LLMs). This access is then used to fraudulently consume expensive computing resources and AI services, leading to substantial cloud charges for the victim organizations.
Unlike cryptojacking, which involves gradual resource consumption, LLMjacking can generate large spikes in fraudulent cloud charges rapidly. Incidents have shown fraudulent charges exceeding $100,000 USD in a single day, with cumulative costs for some organizations reaching $1 million USD before the unauthorized access was detected and stopped.
Methods of Compromise and Monetization
Attackers obtain access through various methods, including phishing, source code leaks, misconfigured environments, or exposed client-side code, which allows them to steal API keys. Another method involves scanning for publicly exposed LLM endpoints. Once initial access is gained, attackers sell this access to services known as "LLM mixers."
These LLM mixers provide their subscribers with unauthorized access to the compromised LLMs. When access is resold, request volumes can spike, sometimes exceeding 1,000 requests per second. The service provider records these requests as legitimate consumption, charging the victim organization for the tokens used.
Widespread Impact Across Sectors
The LLMjacking attacks have a broad geographical footprint, affecting entities in North America, South America, and Asia Pacific. The impacted organizations span various sectors, including financial services, retail, technology, and transportation. GTIG and Mandiant have observed an increase in this type of activity throughout the current year.
Why It Matters
The rapid accumulation of significant fraudulent charges highlights a critical security concern for organizations utilizing LLMs. The financial impact can be substantial, with costs quickly escalating into the hundreds of thousands or even millions of dollars. This necessitates robust security practices around LLM API keys and environment configurations to prevent unauthorized access and mitigate financial losses.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
CVE CVE-2026-402178.8 HIGHHow outlets covered it
Google's Threat Intelligence Group and Mandiant report that LLMjacking attacks are leading to significant fraudulent cloud AI charges, with some organizations losing over $100,000 daily and cumulative costs reaching $1 million. Attackers exploit exposed LLM endpoints or steal API keys, then resell access through 'LLM mixers' that proxy requests, causing victims to incur charges for unauthorized AI usage.
Google has issued a warning about "LLMjacking" attacks, where threat actors hijack enterprise access to Large Language Models (LLMs) to fraudulently consume expensive computing resources. These attacks can lead to rapid, significant cloud charges, with some organizations incurring up to $1 million in costs before detection. The attacks exploit stolen API keys or misconfigured environments to gain unauthorized access to LLM accounts.