Archive — US — techcrunch
GitHub, Microsoft embrace Anthropic’s spec for connecting AI models to datasources
GitHub and Microsoft are joining Anthropic’s MCP steering committee to support a new standard for connecting AI models to data sources, announced at Build 2025. The MCP enables developers to create two-way connections between AI applications and various data sources, increasing functionality in platforms like Microsoft Azure and Windows 11. This move follows similar endorsements by OpenAI and Google, enhancing the integration of AI in development ecosystems.
Key points
- GitHub and Microsoft have joined the MCP steering committee for Anthropic’s standard, enhancing AI model integration with data sources.
- The MCP protocol allows models to access data from diverse business tools and repositories, facilitating task completion.
- MCP enables developers to connect data through 'MCP servers' and create applications ('MCP clients') that communicate with these servers.
- Microsoft plans to offer first-party support for MCP in its platforms, including enhancements in Windows 11 functionalities.
- Microsoft and GitHub will contribute to the MCP standard and aim to provide a secure authorization spec for connecting to MCP servers.
Why it matters
- The adoption of the MCP standard reflects a growing trend towards interoperability and connectivity among AI models and data sources in the IT industry.
- This development aligns with the increased focus on security and data privacy in AI applications as seen with shifts in IT regulations and practices.
- The move also echoes historical collaborations in the tech industry, where major companies converge to standardize protocols (e.g., REST, SOAP) for better software integration.
The collaboration between GitHub, Microsoft, and Anthropic on the MCP standard signifies a pivotal shift towards advanced AI integration, reflecting ongoing trends of interoperability, enhanced developer tools, and an emphasis on security in the evolving IT landscape.