Microsoft's Fedora-Powered Azure Linux 4.0 and Open Source AI Framework Announced

Introduction

At the Open Source Summit this week, Microsoft delivered a series of open source–focused announcements, from updated Linux distributions to new tools for agentic AI. Brendan Burns, corporate vice president for Azure OSS and Cloud Native and co-founder of Kubernetes, outlined the company's shift from cloud-native to an "AI native era" during his keynote. The announcements span both infrastructure and AI, so let's break down the key updates.

Microsoft's Fedora-Powered Azure Linux 4.0 and Open Source AI Framework Announced — Source: itsfoss.com

Azure Linux Updates

Microsoft revealed two significant updates to its Linux offerings aimed at cloud and container workloads.

Azure Linux 4.0: Now in Public Preview

Azure Linux 4.0 is available as a public preview for Azure Virtual Machines. The distro is still under active development, and no downloads are available yet; interested users can sign up for early access via a Microsoft form. Notably, the project's GitHub repository confirms that version 4.0 uses Fedora as its upstream base, with a set of TOML configuration files and targeted overlays applied on top. Packages are pulled directly from Fedora's repositories, with deviations minimized and clearly documented.

Azure Container Linux: Now Generally Available

Azure Container Linux has reached general availability, with a full rollout planned during Microsoft Build on June 2. This is an immutable, container-optimized operating system designed for regulated or security-sensitive environments. By design, it has no package manager and uses a read-only system image, which keeps the attack surface limited while Microsoft manages the entire supply chain end to end. It's particularly suited for teams that need to minimize vulnerabilities in production.

Agentic AI Building Blocks

Microsoft is also pushing forward with open source tools for building and managing AI agents, collectively calling this the "open agentic stack."

Microsoft Agent Framework

The Microsoft Agent Framework is an open source SDK and runtime for multi-agent systems. It consolidates earlier work from Semantic Kernel and AutoGen into a single foundation, making it easier for developers to build coordinated agent workflows.

Agent Governance Toolkit and A2A Protocols

Alongside the framework, Microsoft introduced the Agent Governance Toolkit, which provides identity, policy, and audit controls for AI agent deployments. Additionally, the A2A (agent-to-agent) protocol enables cross-vendor, cross-framework communication between agents, a key step toward interoperability in the growing agent ecosystem.

The Fedora Connection

Microsoft's announcement blog post did not mention Fedora, but the GitHub repository for Azure Linux 4.0 tells a different story. The README explicitly describes Fedora as an "upstream base," and the distro is essentially a set of configuration files applied over Fedora packages. This confirms earlier speculation: in March, we reported that Microsoft was backing a proposal to build x86-64-v3 packages for Fedora 45, motivated by performance gains needed for Azure Linux. Kyle Gospodnetich, a Linux engineer at Microsoft, co-authored that change proposal. There was even talk of Microsoft forking Fedora entirely, but the company chose to work within the Fedora ecosystem instead.

Why the silence about Fedora in the official announcement? Fedora is effectively Red Hat's upstream distribution, and Red Hat is both an Azure partner and a competitor in the enterprise Linux space. Acknowledging the Fedora base would have made for an awkward read in that context.

What This Means

Microsoft's embrace of Fedora as the foundation for Azure Linux 4.0 signals a deeper collaboration within the open source community, even as it navigates competitive tensions with Red Hat. For developers, the new agentic AI tools offer a standardized way to build and govern multi-agent systems. Together, these updates position Microsoft as a serious contributor to open source infrastructure and AI frameworks.

Tags: