Microsoft adopts Google's standards to link AI agents

Microsoft said it is accepting Google's recently launched open protocol to allow AI "agents" to communicate with each other.

On Wednesday, Microsoft announced that it will bring support for Google's Agent2Agent (A2A) specifications to its two AI development platforms, Azure AI Foundry and Copilot Studio. Microsoft also joined the A2A working group on GitHub, contributing to protocols and tools.

“By supporting A2A and building on our open orchestration platform, we set the foundation for the next generation of software – collaborative, observable and designed,” the company wrote in a blog post. “The best agents don’t live in one application or cloud; they will operate across workflows, across models, domains and ecosystems.”

Google's A2A released in early April allows agents (AI-powered semi-autonomous initiatives) to work together on different clouds, applications and services. Using this protocol, the proxy can exchange targets and invoke actions. Developers have obtained a set of interoperable components that can be used to ensure agent collaboration is safe.

Once A2A support reaches Azure AI Foundry and Copilot Studio, a proxy built using the platform will be able to use external proxy clicks for tasks, including using other tools or a proxy hosted externally by Microsoft. For example, Microsoft agents can schedule meetings, while Google agents will draft email invitations.

“(c) Ustamus can build complex multi-agent workflows across internal (agent), partner tools and production infrastructure while maintaining governance and service levels agreements,” the company explained in its blog post. “We are working toward sharing agent agreements with a wider industry.”

While it’s far from perfect, agency technology is attracting more and more investment as businesses seek to increase productivity. According to a recent KPMG survey, 65% of companies are trying AI agents. Markets and market forecasts that the AI ​​agent division will grow from $7.84 billion in 2025 to $52.62 billion in 2030.

TechCrunch Events

Berkeley, CA | June 5

Book now

Microsoft's decision to put weight behind A2A was the standard for Anthropic's standards to connect AI to systems in Copilot Studio with data residency after the company introduced support for MCP. Other major AI model providers, including Google and OpenAI, announced that they will adopt MCP earlier this year.