Threat actors are coming for your AI assets to operationalize their use of AI
Google GTIG reports espionage and crime groups stealing AI models, prompts, and API credentials, plus distillation campaigns and agentic AI attack automation.
Google Threat Intelligence Group's quarterly AI Threat Tracker reports adversaries stealing proprietary models, source code, prompts, and API credentials from government, healthcare, and media targets, including China-based UNC6508 compromising clouds to run unauthorized LLM workloads. Distillation campaigns against Google's models exceeded 100 million prompts launched via thousands of stolen account credentials through proxy networks. Mandiant also observed a financially motivated actor deploy an autonomous multi-agent framework that harvested thousands of third-party credentials in under 6 hours, and a 'Recon' framework on a live C2 server managing over 23,000 stolen credentials including cloud and AI API keys.
- UNC6508 stole AI research and ran LLM infrastructure in compromised clouds
- Distillation campaigns used 100M+ prompts via thousands of stolen credentials
- Autonomous multi-agent framework harvested thousands of credentials in under 6 hours
- Recon framework on live C2 managed 23,000+ credentials including AI API keys
- NSA, FBI, CISA accused China-based labs of industrial-scale model distillation
From the original article
4 min read at csoonline.com
Both state-affiliated cyberespionage group and cybercrime gangs are targeting AI-related documents, configuration files, and proprietary models during intrusions. In addition, the number and scope of distillation attacks, where the knowledge, logic, and reasoning capabilities of LLMs is being extracted with targeted prompts, is increasing.
csoonline.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.