The A2A (Agent2Agent) protocol, created by Google to let AI agents communicate with one another, officially joined the Agentic AI Foundation (AAIF) in August 2026, the Linux Foundation-led body that already hosts Anthropic's Model Context Protocol (MCP). The shift brings, for the first time, the two most widely used building blocks of the agentic ecosystem under the same neutral roof.
The move is no small matter for AI agent makers: until now, each provider could evolve its own communication standard at its own pace, risking a fragmented market. The convergence of Google, Anthropic, OpenAI, Microsoft, Amazon, and Block around a shared governance structure changes the equation.
MCP and A2A, two complementary protocols
Contents
The two standards don't play the same role. MCP, which Anthropic contributed to the foundation as soon as the AAIF launched in December 2025, standardizes how an agent accesses its tools and data: internal databases, search engines, business APIs. The protocol was already described in the new Anthropic standard designed to connect agents to physical machines.
A2A, for its part, governs agent-to-agent communication: how one agent delegates a task to another, negotiates a result, or coordinates a workflow with agents published by competing vendors. This distinction is crucial for cross-framework interoperability use cases, a topic already covered on this site regarding the connection of CrewAI to companies' internal databases.
Rapid growth in under a year
Launched in December 2025 with fewer than 40 member organizations, the AAIF now claims more than 250, including Google, Microsoft, Amazon, Anthropic, OpenAI, Bloomberg, Shopify, and Block. The A2A protocol, open-sourced by Google Cloud in April 2025 and then transferred to the Linux Foundation in June of that same year, thus reaches a new milestone by formally joining the AAIF's shared governance rather than remaining tied to a single player.
- December 2025: the Linux Foundation launches the AAIF, with MCP (Anthropic), goose (Block), and AGENTS.md (OpenAI) as its first projects
- April 2025: Google Cloud open-sources the A2A protocol
- August 2026: A2A formally joins the AAIF alongside MCP
- More than 250 member organizations, up from fewer than 40 at launch
What this means for companies deploying agents
For companies building AI agent workflows, the promise is concrete: being able to mix agents and tools from different providers without having to code a custom integration for every pair. That's exactly the logic at work in agentic commerce, where a shopping agent needs to be able to talk to both a merchant's systems and a payment provider's, regardless of which vendor built each piece.
Still, neutral governance alone doesn't guarantee uniform adoption: each vendor keeps its own proprietary extensions on top of the shared protocols, and the history of software standards shows that multiple implementations often coexist, even under the same umbrella. The challenge over the coming months will be to verify whether agents certified as A2A- and MCP-compatible actually work together in production, beyond the demos.
Frequently Asked Questions
What's the difference between MCP and A2A?
MCP standardizes how an agent connects to its tools and data (databases, APIs, search), while A2A governs direct communication between two independent agents, including ones from different vendors.
Who governs these protocols as of 2026?
Both standards are hosted by the Agentic AI Foundation, a Linux Foundation structure launched in December 2025 by Google, Anthropic, OpenAI, Microsoft, Amazon, and Block, which now counts more than 250 members.
Why does this unification benefit companies?
It reduces the need to build a custom integration for every combination of agents and tools, making it easier to deploy agent chains that span multiple providers.
Sources
Axios —
Linux Foundation (official press release) —
Techzine Global
Author Profile
Latest entries
Planet15 November 2026CO2 Emissions: Why Transportation Remains France's Climate Blind Spot
Artificial Intelligence12 November 2026Imagine Enterprise: IBM and Microsoft Team Up on AI Agents for Large Businesses
Economy11 November 2026Livret A Rate: Understanding How It's Calculated and Its Limits
Shopping8 November 2026Black Friday: Key Dates and the Budget French Shoppers Are Planning
This article was written with the help of artificial intelligence. Editorial policy









