Google Cloud has introduced Agent Compute Engine, a purpose-built service designed to run autonomous AI agents with hardware-isolated security. The new infrastructure uses microVMs that boot rapidly and pause when idle to improve efficiency for high-churn agent sessions. Alongside this launch, the company is extending its agent governance capabilities to cover agents running across various cloud and on-premises environments. These updates aim to provide organizations with better control, observability, and security for managing large fleets of AI agents.
Read the full story at cloud.google.com.
