NEW ORLEANS – Microsoft has joined more than 50 major industry businesses (including Salesforce, Oracle, and SAP) in support of Google's Agent2Agent (A2A) open‑protocol, a system specifically designed for communication between AI agents.
“By supporting A2A and building on our open orchestration platform, we’re laying the foundation for the next generation of software — collaborative, observable, and adaptive by design,” Microsoft shared in a blog post. “The best agents won’t live in one app or cloud; they’ll operate in the flow of work, spanning models, domains, and ecosystems.”
AI Agents and A2A
An AI agent is a software program that can autonomously detect information, make decisions, and take actions to achieve specific goals. Google’s A2A protocol enables AI agents like chatbots and virtual assistants to communicate and interact seamlessly with other apps and services by defining standardized message formats and capabilities.
Despite being competitors, the big tech companies are finding it more beneficial to cooperate on AI given the rate of change and uptake. ChatGPT, for example, reached 1 million users in just 5 days.
AI agents often exist in silos, unable to collaborate easily across different systems. A2A breaks down those walls so that AI tools can cooperate and share tasks and this helps developers integrate AI capabilities into their apps without having to build custom connections for each different agent ecosystem.
The result is that the largest tech companies are caught between protecting the money they make from their current businesses and spending on new AI technology, even if the new technology risks hurting their old ways of generating revenue.
“Tech giants like Google and Microsoft face the innovator's dilemma as they balance legacy revenue with disruptive AI investments,” said Henry Hays, CEO and co-founder of DisruptREADY, a local business intelligence firm specializing in teaching businesses how to use emerging technology.
Efficiencies and Layoffs
Microsoft’s recent laying off of 7,000 people happened not because they were doing a bad job, but because new AI technology can do their jobs more efficiently. As AI agents become more advanced and widely deployed, similar shifts are likely across a broad range of industries, reshaping how tasks are performed and reducing the need for certain roles.
“Companies leveraging AI, exemplified by Lodestar's recent payroll processing time reduction from 3 days to 30 minutes, gives businesses significant competitive advantages,” said Hays. Lodestar is a Louisiana‑based outsourced CFO firm that provides cloud-based payroll and HR solutions to streamline businesses’ employee compensation and benefits processing.
“DisruptREADY helped us transform a cumbersome, Excel‑heavy payroll process into a streamlined, automated solution. … We’ve seen immediate value and are excited to continue exploring new ways to work with them.”
According to a recent KPMG survey, 65% of companies are experimenting with AI agents. Markets and Markets projects that the AI agent segment will grow from $7.84 billion in 2025 to $52.62 billion by 2030.
“Businesses ignoring AI adoption risk facing steep costs and efficiency disadvantages compared to competitors utilizing AI technologies.”
Microsoft introduced their new AI Agent functionality using A2A technology to a private group of developers and businesses at their Microsoft Build 2025 Annual Developer Conference in May and indicated that the release date is imminent. At the conference, Microsoft demonstrated how AI Agent functionality enables structured communication, exchanges goals, invokes actions, and returns results across cloud and platform boundaries.
Microsoft has already announced plans to integrate A2A into its Azure AI Foundry and Copilot Studio platforms. Azure helps businesses develop and run custom AI systems in the cloud, while Copilot Studio lets companies build and manage virtual assistants and chatbots for everyday tasks.
Integrating Other Protocols
Microsoft has also decided to support Anthropic’s MCP (Message Coordination Protocol), which helps AI agents plan and coordinate within a single conversation, managing multi-step reasoning and tool use. This would complement A2A in building complex, multi-agent applications.
