Microsoft Backs Google's Open Agent2Agent Protocol to Power Multi-Agent Collaboration
Microsoft is expanding support for Google’s open AI agent interoperability protocol with the upcoming integration of Agent2Agent (A2A) in Azure AI Foundry and Copilot Studio. This move enables AI agents to collaborate across platforms, clouds, and organisations.
Enabling Interoperability
The company said the A2A protocol enables “structured agent communication—exchanging goals, managing state, invoking actions, and returning results securely and observably”.
“Open protocols like A2A and MCP are key to enabling the agentic web. With A2A support coming to Copilot Studio and Foundry, customers can build agentic systems that interoperate by design,” Microsoft CEO Satya Nadella said.
The protocol supports popular tools such as Semantic Kernel and LangChain, and routes all communications through Microsoft’s enterprise safeguards, including Entra, mutual TLS, Azure AI Content Safety, and full audit logging.
Enhanced Capabilities
“With support for A2A, Azure AI Foundry customers can build complex, multi-agent workflows that span internal copilots, partner tools, and production infrastructure—while maintaining governance and SLAs (Service Level Agreements),” the company said in a blog post. “Copilot Studio agents will be able to securely invoke external agents, including those built with other platforms or hosted outside Microsoft.”
Microsoft mentioned that over 10,000 organizations have already adopted its Agent Service in the past four months. At least 2.3 lakh organizations have used Copilot Studio, including 90% of the Fortune 500.
Future Contributions
Microsoft has joined the A2A working group on GitHub and plans to contribute to the protocol’s specification and tooling. A public preview of A2A support in Foundry and Copilot Studio is expected soon.
“We’re laying the foundation for the next generation of software—collaborative, observable, and adaptive by design,” the company said. “The best agents won’t live in one app or cloud; they’ll operate in the flow of work, spanning models, domains, and ecosystems.”
Agentic Computing Revolution
The company sees this development as part of a larger trend in what it calls “agentic computing”, which it says marks a fundamental change in how software is developed and deployed.
Join us for “The Optimized AI Stack,” a session on deploying LLMs on AI PCs using CPU, GPU, and NPU acceleration with Intel’s OpenVINO toolkit.
The Data Engineering Summit (DES) 2025, presented by AIM Media House, returns for its fourth edition on May 15–16.
Email:
[email protected]
Our Offices
AIM India
1st Floor, Sakti Statesman, Marathahalli – Sarjapur Outer Ring Rd, Green Glen Layout, Bellandur, Bengaluru, Karnataka 560103
AIM Americas
99 South Almaden Blvd. Suite 600 San Jose California 95113 USA
© Analytics India Magazine Pvt Ltd & AIM Media House LLC 2025
