Cybercriminals are increasingly targeting corporate AI accounts through "LLMjacking," using stolen API keys and session tokens to exploit expensive compute resources for illicit, large-scale operations.
Key Points
- Hackers use infostealer malware to siphon credentials, allowing them to resell access to major AI platforms like OpenAI, Anthropic, and Google at significant discounts.
- CrowdStrike reported an 89% increase in AI-related adversary activity between July 2025 and June 2026.
- A September 2026 analysis by Okta identified hundreds of unexpired AI service tokens within a single 7 GB dump of stolen logs.
- Autonomous AI-agent campaigns compromised infrastructure across 395 organizations in 48 countries during September 2026.
- Attackers often evade detection by throttling usage to mimic legitimate traffic patterns, making unauthorized activity difficult for traditional security systems to identify.