Google to Adopt Anthropic's AI Data Connection Standard

Google to Integrate Anthropic’s Model Context Protocol into Gemini

Following a recent announcement by OpenAI regarding the adoption of Anthropic’s Model Context Protocol (MCP), Google is now also implementing support for this standard.

Demis Hassabis, CEO of Google DeepMind, revealed via a post on X (formerly Twitter) on Wednesday that Google’s Gemini models and Software Development Kit (SDK) will gain compatibility with MCP.

The Significance of MCP

Hassabis characterized MCP as a valuable protocol, noting its increasing recognition as an open standard crucial for the evolving landscape of AI agents.

He expressed enthusiasm for collaborative development, stating a desire to further refine MCP alongside the Anthropic team and other industry stakeholders.

How Model Context Protocol Functions

MCP facilitates the ability of AI models to access and utilize data from various sources.

These sources include essential business applications, specialized software, content repositories, and application development environments.

The protocol empowers developers to establish bidirectional communication between data sources and AI-driven applications, such as conversational AI chatbots.

Building with MCP: Servers and Clients

Developers can make data accessible through dedicated “MCP servers.”

Correspondingly, they can create “MCP clients” – encompassing applications and workflows – that connect to these servers upon request.

Growing Industry Adoption

Since Anthropic released MCP as an open-source project, a growing number of companies have begun integrating it into their platforms.

These companies include Block, Apollo, Replit, Codeium, and Sourcegraph, demonstrating the protocol’s increasing relevance and utility.

This widespread adoption underscores MCP’s potential to become a foundational element in the future of AI integration.