Google Joins Anthropic's MCP Movement, Boosting AI Data Integration Standards

AI models to data

In a major step toward streamlining how AI interacts with real-world data, Google is now embracing Anthropic’s Model Context Protocol (MCP)—a move that signals growing industry alignment around a common standard for AI-to-data connectivity.

Announced on Wednesday via a post on X (formerly Twitter) by Demis Hassabis, CEO of Google DeepMind, the tech giant confirmed that it will add support for MCP to its Gemini AI models and SDK. While Hassabis didn’t share a timeline, his message made it clear: MCP is quickly becoming the default language for AI models to access external data sources.

“MCP is a good protocol and it’s rapidly becoming an open standard for the AI agentic era,” Hassabis wrote. “Looking forward to developing it further with the MCP team and others in the industry.”

What is MCP?

Model Context Protocol (MCP) allows AI models to securely interact with data across various platforms—like business software, content repositories, developer environments, and more. By establishing two-way communication between data sources and AI tools, MCP enables smarter and more responsive AI systems.

Developers can now:

• Set up MCP servers to expose real-time data.

• Build MCP clients—applications, workflows, or chatbots—that dynamically connect to and act on that data.

It’s an important evolution in making AI not just reactive but context-aware, especially in enterprise environments.

Rapid Industry Adoption

Google’s adoption of MCP follows closely behind OpenAI, which also pledged support for the standard in a surprising move that underscored Anthropic’s growing influence. Since Anthropic open-sourced MCP, a growing list of tech companies have joined the fold, including:

• Block

• Apollo

• Replit

• Codeium

• Sourcegraph

The momentum suggests that MCP could become the de facto integration standard for AI agents in everything from code completion tools to enterprise chatbots.

Why This Matters

The embrace of a common data protocol is a huge leap forward for developers building in the AI space. Instead of wrestling with custom APIs or disconnected tools, they’ll be able to rely on a consistent method for securely and efficiently connecting AI to the data it needs—in real-time and across platforms.

As the "AI agentic era" unfolds, protocols like MCP will be the glue binding intelligent models to the ecosystems that make them useful.