Google is the latest tech giant to throw its weight behind Anthropic’s Model Context Protocol, a rising industry standard designed to help AI models interact more effectively with external data sources.
In a post on social media platform X on Wednesday, Google DeepMind CEO Demis Hassabis announced that support for MCP would be integrated into Google’s Gemini AI models and software development kit (SDK). While no specific timeline was provided, the move positions Google alongside OpenAI, which recently committed to adopting the protocol.
“MCP is a good protocol and it’s rapidly becoming an open standard for the AI agentic era,” Hassabis said. “We look forward to developing it further with the MCP team and others in the industry.”
Originally introduced by Anthropic, MCP allows developers to create two-way connections between AI models and a wide range of data sources — from enterprise tools and cloud software to content libraries and app development environments. The goal is to enable AI systems to not just respond to queries, but also complete complex tasks by pulling in real-time data and interacting with external applications.
The protocol uses a client-server model in which “MCP servers” host data or services, and “MCP clients” — such as apps or workflows — request access to them dynamically. This structure allows for more flexible and secure integration of AI agents into real-world business operations.
Since Anthropic open-sourced the protocol earlier this year, several tech companies including Block, Apollo, Replit, Codeium, and Sourcegraph have adopted it across their platforms. Industry analysts say growing support for MCP marks a step toward standardizing how AI models interface with the tools and information they need to operate in real-world environments.
