This article outlines various security risks associated with AI agents and their infrastructure, including issues like chat history exfiltration and prompt injection. It emphasizes the need for a comprehensive security platform to monitor and govern AI operations effectively.

Security researchers found serious vulnerabilities in Ollama and NVIDIA Triton Inference Server that could allow remote code execution. Although these flaws have been patched, they highlight growing security concerns around AI infrastructure and the shift in focus from model exploitation to infrastructure vulnerabilities.

State-sponsored hackers are increasingly exploiting vulnerabilities in critical infrastructure systems, particularly targeting sectors such as energy and transportation. These attacks are becoming more sophisticated and coordinated, posing significant risks to national security and public safety. Governments are urged to enhance their cybersecurity measures to mitigate these threats effectively.