Kong AI Gateway 2.0 on Google Cloud: Securing GKE, Cloud Run, and Vertex AI

Kong has announced the release of Kong AI Gateway 2.0, designed to address the complexities of securing AI-native ecosystems across Google Cloud. As enterprises increasingly deploy workloads across Google Kubernetes Engine (GKE) and Cloud Run, managing AI traffic—including LLM calls and Model Context Protocol (MCP) servers—requires more than traditional API management. The 2.0 update introduces dedicated abstractions for AI model providers, agents, and MCP servers, enabling unified authentication, observability, and cost governance. By leveraging Google Cloud Workload Identity, the gateway eliminates the need for static service account keys when interacting with Vertex AI. It also provides advanced features such as token-based rate limiting, semantic prompt inspection, and agent-to-agent security. This release allows organizations to maintain a consistent security posture across diverse compute environments, ensuring that AI-driven applications remain protected, performant, and compliant within a fragmented infrastructure.