Google Maps Platform

Google Maps Platform provides mapping, routing and places data through APIs, and its current direction is defined by connecting that data to AI systems rather than by extending the mapping features themselves. Several capabilities point the same way, and together they describe a specific bet about how location data will be consumed.

Grounding with Google Maps improves AI responses in Gemini, AI Studio and the Agent Platform, while Grounding Lite brings fresh Maps data to any large language model through the Model Context Protocol. MCP support is the more consequential of the two, because it makes Maps data available to AI systems that are not Google's, which is a different commercial posture from keeping the data inside one assistant.

Grounding addresses a real failure mode rather than a hypothetical one. Language models produce confident answers about places, opening hours and routes from training data that was already stale when the model was built, and a user cannot tell a remembered answer from a current one. Connecting the model to live Maps data replaces recall with lookup.

The Maps Agentic UI Toolkit, currently experimental, builds interactive map experiences into agentic applications, and Maps Imagery Grounding in private preview generates media grounded in physical reality. Place, area and review summaries provide AI-generated descriptions through Place Details, which is a useful shortcut where an application needs a readable description of somewhere rather than a list of attributes.

Alongside the AI direction, Google Earth combines imagery with professional geospatial data, no-code tools and Gemini capabilities, and Geospatial Analytics provides datasets with AI-powered analysis tools. The no-code element matters because geospatial analysis has historically required specialist GIS skills, and the analysts who most need the answers rarely have them.

Pricing follows Google Cloud consumption per API call rather than published plan tiers, with a monthly free allowance.