Microsoft goes all in on Anthropic's MCP standard for safer AI agent deployments
Briefly

As AI technology gains traction, Microsoft recently joined the growing support for the Model Context Protocol (MCP), an AI data connection standard developed by Anthropic. Announced during the Build developer event, Microsoft's endorsement involves integrating MCP in platforms like GitHub, Azure, and Windows 11. However, Microsoft emphasizes the importance of security, citing risks linked to untrusted input data and potential vulnerabilities in AI systems. The initiative aims to foster secure, scalable adoption of MCP as developers create AI applications across its platforms.
"From a security perspective, the input and training data for a LLM are considered untrusted."
Read at ZDNET
[
|
]