Google Agent2Agent protocol joins Linux Foundation to enable cross-cloud interoperability
Google has donated its Agent2Agent (A2A) open protocol to the Agentic AI Foundation under the Linux Foundation. The standard enables AI agents to discover and delegate tasks across different frameworks and cloud providers, aiming to reduce vendor lock-in for enterprise AI systems.
Key Takeaways
- A2A enables agents to publish structured 'agent cards' for automated discovery and task delegation across vendor boundaries
- Founding members of the Agentic AI Foundation include AWS, Microsoft, Salesforce, ServiceNow, and Cisco
- Huawei has standardized A2A for its Celia assistant to communicate with in-app agents on HarmonyOS
- IBM merged its Agent Communication Protocol into A2A in August 2025 to support a single industry standard
Why It Matters
The donation of A2A to a neutral foundation signals a shift toward an open agentic ecosystem where enterprises can mix and match AI services based on performance rather than infrastructure constraints. For the streaming and media sectors, this interoperability allows specialized agents—such as those for content recommendation or metadata tagging—to collaborate across siloed platforms like AWS Bedrock and Azure AI Foundry. By standardizing the 'agent-to-agent' layer, the industry avoids the fragmentation that previously slowed cross-platform media workflows. Watch for whether Tencent’s WeChat integration with Huawei assistants triggers a broader wave of A2A adoption among Western consumer application developers.
Additional Context
The Agentic AI Foundation under the Linux Foundation is rapidly accumulating governance over competing agent communication standards. In July 2025, Ericsson published a detailed framework positioning agentic AI as the pathway to autonomous network level 5, claiming an 80 percent reduction in time spent on analysis and decision-making processes when agentic systems handle network operations. That telecom use case illustrates the kind of cross-vendor orchestration the Google Agent2Agent protocol is designed to standardize: agents from different vendors discovering each other's capabilities and delegating tasks without bespoke integration layers. The Linux Foundation's neutral governance model mirrors how Kubernetes achieved multi-cloud portability, and the addition of A2A alongside existing projects like Model Context Protocol and goose gives the foundation a broader surface area for enterprise adoption.
On the competitive front, Blue Planet and Telefónica Deutschland completed a joint proof of concept using agentic AI to power 5G network slicing services, demonstrating that AI agents can reduce slice specification design from weeks to minutes. The PoC integrated Blue Planet AI Studio with Telefónica's multi-domain service orchestration environment, validating intent-based approaches for autonomous operations. ABI Research has forecast network slicing to become a $19.5 billion market by 2028, and agentic orchestration standards like A2A could accelerate that timeline by enabling agents from different vendors to coordinate slice lifecycle management without proprietary glue code. Meanwhile, Telefónica Germany selected Blue Planet in May 2025 to drive cloudification and autonomous network transformation, deploying Inventory, Orchestration, and 5G Network Slicing applications across radio access, transport, and core domains.
Ericsson's own AI-RAN work provides a concrete benchmark for what agentic optimization delivers at the infrastructure layer. At MWC 2026, Ericsson's networks chief Per Narvinger detailed how AI models improve spectrum efficiency by 10 percent over algorithms refined for 30 years, with Bell Canada running the first field tests in April 2025 and AT&T following at MWC on Intel-based cloud RAN. Narvinger noted that by end of 2026, Ericsson will have 10 AI-ready radio models with embedded neural accelerators, up from one currently in market. The Ericsson Mobility Report separately quantified that gen AI traffic currently represents only 0.06 percent of total network data but shifts the uplink-downlink ratio to 26-74, compared to the typical 10-90 split. For streaming operators planning agent-driven content workflows, these traffic pattern shifts will influence capacity planning as A2A-enabled agents increasingly coordinate video processing, metadata enrichment, and delivery optimization across distributed cloud environments.
Read full article at aimagazine.com
Enjoy our coverage?
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source