
The rush to stay ahead in the AI era has led to unprecedented cooperation between traditional rivals at both business and governmental levels. While companies are increasingly working together on dedicated projects that benefit their joint customers, there is still a disconnect when it comes to widespread interoperability.
That may be about to change as OpenAI and Microsoft become the latest organizations to throw their support behind the Model Context Protocol (MCP), an open standard first floated by Anthropic, that could have a profound impact on how AI agents interact across various tools and environments. Developers have recently unveiled the latest specifications for MCP and, with renewed support, it could open the door for widespread agentic AI deployment.

AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
What is MCP?
Before we discuss the latest announcement regarding MCP, let’s recap. Anthropic introduced MCP in 2023 as an open-source method for standardizing how data sources connect in AI use cases. The recent updates to the protocol position MCP to become the preferred standard for AI agent connectivity.
The upgrades to MCP focus on AI agent security, functionality, and interoperability. They include the introduction of the OAuth 2.1-based authorization framework for secure communication between agents and servers.
Streamable HTTP transport supports bidirectional data flow in real-time and enhanced compatibility, while latency between agents and tools has been reduced through JSON-RPC request batching. Finally, new tool annotations enable AI agents to carry out more complex reasoning actions with access to richer metadata.
What’s the Deal with OpenAI and Microsoft?
Support for MCP from OpenAI was confirmed by the company’s CEO Sam Altman in a message on X, reading, “people love MCP and we are excited to add support across our products. available today in the agents SDK and support for chatgpt desktop app + responses api coming soon!”
Although brief, this message is significant. The company behind the world’s most popular AI platform is ultimately adopting a protocol developed and introduced by a rival company to support interoperability. And, incredibly, it isn’t the only one.
Microsoft has also voiced its support for MCP, including the release of Playwright-MCP, a “Model Context Protocol (MCP) server that provides browser automation capabilities using Playwright.” Essentially, the new server enables AI agents to interact with web pages instead of simply answering questions about them.
Closing Thoughts
The news that OpenAI and Microsoft are supporting MCP is significant. Anthropic remains a major rival, but the benefit to the broader AI ecosystem seems to be taking precedence over competition. This rapidly evolving ecosystem continues to produce unprecedented scenarios.
Interoperability is crucial. As AI agents present new opportunities — especially interactive roles in workflows — companies that don’t collaborate risk falling behind.
It’s promising to see what may become a universal AI agent protocol. Hopefully, this level of interoperability also supports shared values and leads to governance guidelines driven by the same companies adopting the standards.
Ask Cloud Wars AI Agent about this analysis