Google has introduced fully managed Model Context Protocol (MCP) servers to simplify how AI agents connect to real-world tools and enterprise data. The rollout is designed to eliminate the fragile, manual integrations developers currently rely on, offering direct, secure access to Google services such as Maps, BigQuery, Compute Engine, and Kubernetes Engine. The launch follows the debut of Google’s latest Gemini 3 model, pairing improved reasoning capabilities with more dependable tool interoperability for production-grade AI workflows.
The managed MCP endpoints allow developers to activate integrations with a simple URL, reducing setup time from weeks to minutes. Early use cases include analytics agents querying BigQuery directly and operational agents interacting with cloud infrastructure. For location-based AI tools, grounding agents in live Google Maps data enables more accurate and up-to-date planning.
Initially released in public preview at no additional cost for existing enterprise customers, the MCP servers are protected through Cloud IAM, Model Armor, and full audit logging to ensure governance and security. Google plans to extend MCP coverage across storage, databases, monitoring, and security services in the coming months.
