Just weeks after OpenAI committed to using Anthropic’s Model Context Protocol (MCP), Google is now following the same path. The company has announced plans to integrate MCP into its Gemini AI models and SDK, signaling growing industry-wide support for the emerging open standard.
The update came directly from Google DeepMind CEO Demis Hassabis, who shared the news in a post on X this Wednesday. While he didn’t offer a specific timeline, Hassabis made it clear that Google sees MCP as a critical piece of the puzzle for AI’s future.
“MCP is a good protocol and it’s rapidly becoming an open standard for the AI agentic era,” Hassabis wrote. “Look forward to developing it further with the MCP team and others in the industry.”
Originally developed by Anthropic, the Model Context Protocol allows AI models to access external data from business tools, apps, and content systems. This capability is vital for enabling real-world functionality — like AI assistants that can pull live data or perform tasks based on current context.
With MCP, developers can set up “MCP servers” that act as gateways to data, and create “MCP clients” — apps or workflows — that request and use that data. These two-way connections give AI models a deeper, real-time understanding of the systems they interact with.
Since MCP was open-sourced, a wave of support has followed. Major players including Block, Apollo, Replit, Codeium, and Sourcegraph have already adopted the protocol, integrating it into their platforms to unlock smarter, context-aware AI capabilities.
Google’s adoption of MCP marks a significant step forward in standardizing how AI models communicate with their environments. As more tech leaders align on shared infrastructure like MCP, the vision of truly capable, integrated AI agents becomes more tangible — and much closer to reality.
