
Image: ClaudeMCP
MCP, or Model Context Protocol, is an open-source specification developed by Anthropic, the creator of the AI assistant Claude. This protocol is designed to establish a universal interface that enables AI models to connect with external tools, data sources, and systems—facilitating seamless integration for application developers seeking to enhance the interoperability of their AI-powered services.
By standardizing how AI models access external resources—ranging from local files and databases to enterprise applications and internet-based services—MCP significantly enhances both the utility and adaptability of AI systems.
Previously, OpenAI announced that its Agent SDK would support the MCP protocol. Since then, numerous developer platforms and services have followed suit. Now, Google joins the initiative by announcing MCP compatibility within its Gemini AI ecosystem, offering developers a streamlined way to leverage external resources through the protocol.
According to Google DeepMind engineers, the team plans to integrate MCP support into both Gemini models and the associated SDK. Moreover, DeepMind has expressed a strong interest in collaborating with the MCP community and industry partners to further refine the protocol and expand its impact.
Interestingly, Google CEO Sundar Pichai hinted at internal deliberations regarding MCP support during remarks on March 31—signaling the company’s early awareness of the protocol. Just ten days later, Google officially confirmed that Gemini would support MCP.
Given the current momentum, MCP appears poised to become a de facto standard in the AI industry. With major players embracing the protocol, it is likely that many more AI providers and application developers will soon adopt MCP—paving the way for more accessible, integrated, and user-friendly AI experiences.