The Protocol Landscape Takes Shape
A new axis of competition in enterprise artificial intelligence (AI) infrastructure has emerged around which open standard will govern how autonomous AI agents communicate, discover one another, and coordinate tasks across organizational boundaries. Two protocols now dominate that conversation: Anthropic's Model Context Protocol (MCP), introduced in November 2024, and Google's Agent2Agent (A2A) protocol, launched in April 2025.
<cite index="11-17">The Model Context Protocol (MCP) is an open standard and open-source framework introduced by Anthropic in November 2024 to standardize the way artificial intelligence (AI) systems like large language models (LLMs) integrate and share data with external tools, systems, and data sources.</cite> <cite index="9-4">The architecture allows developers to either expose their data through MCP servers or build AI applications — MCP clients — that connect to those servers.</cite> <cite index="10-6">Since launching MCP in November 2024, adoption has been rapid: the community has built thousands of MCP servers, SDKs are available for all major programming languages, and the industry has adopted MCP as the de-facto standard for connecting agents to tools and data.</cite>
Google Launches A2A, Microsoft Follows
<cite index="2-2">Google Cloud launched the Agent2Agent Protocol (A2A), an open standard designed to enable AI agents to collaborate across enterprise systems regardless of their underlying frameworks or vendors.</cite> <cite index="7-4">Google's official announcement went out on April 9, 2025, at Google Cloud Next.</cite> <cite index="7-5">More than 50 launch partners came along, including Accenture, Atlassian, Box, Cohere, Deloitte, Elastic, LangChain, MongoDB, PayPal, Salesforce, SAP, ServiceNow, and UiPath.</cite>
<cite index="32-1">On May 7, 2025, Microsoft announced its support for the open Agent2Agent (A2A) protocol, marking a significant step towards fostering interoperability among AI agents across diverse platforms and ecosystems.</cite> In its official blog post, <cite index="29-3">Microsoft stated it is committed to advancing open protocols like A2A, coming soon to Azure AI Foundry and Copilot Studio, which will enable agents to collaborate across clouds, platforms, and organizational boundaries.</cite> <cite index="3-18">Microsoft also joined the A2A working group on GitHub and plans to contribute to the spec and tooling.</cite>
Linux Foundation Takes Stewardship
<cite index="19-1">At Open Source Summit North America on June 23, 2025, the Linux Foundation announced the launch of the Agent2Agent (A2A) project, an open protocol created by Google for secure agent-to-agent communication and collaboration.</cite> <cite index="20-6">The founding members of the Linux Foundation project were Amazon Web Services, Cisco, Google, Microsoft, Salesforce, SAP, and ServiceNow.</cite> Salesforce, whose Agentforce platform is a major enterprise agent product, stated at the time: <cite index="19-21,19-22,19-23">"At Salesforce, we believe the future of enterprise AI lies in seamless agent-to-agent collaboration. That's why we are excited to be part of the Linux Foundation's A2A Project."</cite>
Microsoft's VP of Product for Azure AI Foundry added: <cite index="27-22,27-23">"Microsoft is committed to shaping the future of agentic AI by combining open interoperability with the enterprise-grade capabilities that organizations need to deploy agents responsibly and at scale. We welcome the announcement of A2A as a neutral nonprofit project."</cite>
A2A and MCP: Complementary, Not Competing
Industry observers and the companies themselves have framed the two protocols as operating at different layers rather than as direct rivals. <cite index="28-4,28-5,28-6">A2A is described as complementary to the Model Context Protocol (MCP), another Linux Foundation project. A2A defines how agents communicate and coordinate with each other across organizational boundaries, while MCP defines how agents connect to internal tools and data sources. Together, they form a foundational layer for interoperable, multi-agent systems.</cite>
<cite index="11-25">In December 2025, Anthropic donated MCP to the Agentic AI Foundation (AAIF), a directed fund under the Linux Foundation, co-founded by Anthropic, Block, and OpenAI, with support from other companies.</cite> <cite index="18-23">Salesforce announced in June 2025 that Agentforce 3 also supports MCP as interoperability plays a crucial role in the company's overall AI strategy.</cite>
Scale and Adoption
<cite index="28-1">Since April 2025, the number of organizations supporting A2A has grown from more than 50 to over 150 — including AWS, Cisco, Google, IBM, Microsoft, Salesforce, SAP, and ServiceNow.</cite> <cite index="8-3">Microsoft, AWS, Salesforce, SAP, and ServiceNow are running A2A in production environments.</cite> On the MCP side, <cite index="12-8">Model Context Protocol enterprise adoption among Fortune 500 companies has reached a 28% implementation rate in less than 18 months, with enterprises deploying MCP servers for production AI workflows.</cite>
With both protocols now under neutral Linux Foundation governance and backed by overlapping coalitions of incumbents, the 2025–2026 period is solidifying the agent infrastructure stack — though enterprise procurement decisions, security models, and pricing remain unresolved variables in broad deployment.