Google has transferred its Agent2Agent (A2A) protocol, designed for AI agent communication, to the Linux Foundation to foster open-source collaboration. The protocol, previously adopted by Microsoft Azure, has now gained backing from Amazon Web Services (AWS), aiming to enhance interoperability among AI agents while maintaining neutrality.
The A2A protocol addresses the need for seamless AI agent interaction across different applications and systems, circumventing vendor-specific barriers. This initiative comes amidst similar efforts like Anthropic's Model Context Protocol (MCP), which targets communication between AI agents and external databases. Microsoft supports both protocols, advocating for broader adoption of open standards within the AI industry.
The Linux Foundation recently announced the creation of the Agent2Agent project at the North American Open Source Summit, involving major industry stakeholders, including AWS, Cisco, Google, Microsoft Azure, Salesforce, SAP, and ServiceNow. The project aims to develop an open, interoperable ecosystem for AI agents, preventing the development of isolated vendor-specific AI environments.
This ecosystem relies on the A2A protocol, contributed by Google, that establishes a shared language enabling AI agents to identify capabilities, exchange information securely, and coordinate complex tasks. More than 100 companies have now endorsed the protocol, with AWS and Cisco among the latest official participants.
Jim Zemlin, executive director of the Linux Foundation, highlighted that housing A2A under the Linux Foundation ensures neutral governance, cooperation, and long-term sustainability. This move is intended to foster a new phase of agent collaboration and productivity improvements.
AWS leadership emphasized the protocol's potential to enhance customer experience through agent-based AI. Swami Sivasubramanian, vice president of Agent-Based AI at AWS, described agent-based AI as integral to improving nearly all customer-facing processes. AWS plans to support the project by contributing relevant frameworks, protocols, and services.
Cisco is actively implementing A2A within its open-source AGNTCY initiative to establish interoperability layers that extend across enterprise boundaries, according to Vijoy Pandey, General Manager and senior vice president of Cisco Outshift. The emphasis is on facilitating AI agent communication across disparate systems and organizations.
From Microsoft's perspective, open standards like A2A are crucial, but enterprise-grade capabilities are also necessary for responsible and scalable AI agent deployment. Yina Arenas, vice president of Azure AI Foundry Products, expressed Microsoft's commitment to combining open interoperability with robust enterprise functionalities. She welcomed the launch of the neutral, nonprofit A2A project and indicated enthusiasm for collaboration to advance agent development standards.
The transfer of A2A to the Linux Foundation and the growing support from major cloud providers signal an industry recognition of the importance of open, interoperable AI agent communication. As AI agents become more prevalent across industries, protocols like A2A are expected to play a foundational role in enabling secure, efficient coordination among diverse AI systems.
Article edited by Jack Wu
