Microsoft Ignite Special Edition: Agent 365, Claude on Azure, and a Record DDoS Attack

This Week in Cloud — November 20, 2025

Welcome back to The Cloud Cover, your essential guide to navigating the dynamic world of cloud for Solutions Architects, engineers, and IT leaders. Autonomous agents are gaining momentum, providers are racing to strengthen governance layers, and a series of outages reminded everyone how fragile the stack still is. Let’s unpack the week.

If Microsoft Ignite 2025 had a single thesis, it was this: The era of the passive Copilot is ending, and the era of the autonomous Agent has begun. But Microsoft’s most strategic move wasn’t just launching new agents; it was positioning itself to govern them. With the introduction of Agent 365, Microsoft is attempting to solve the looming "Shadow AI" crisis before it fully explodes. 

Agent 365 acts as a control plane for the "silicon workforce." As enterprises start deploying thousands of autonomous software entities—some built on Azure, some open-source, some "rogue"—CIOs are facing a nightmare of unmonitored decision-making. Microsoft’s solution is a centralized registry that offers identity verification (Entra Agent ID), visual observability to map execution flows, and "kill switches" via Defender to terminate agents acting anomalously. 

The implication here is massive. By supporting interoperability with third-party agents (like those from ServiceNow or Databricks), Microsoft is acknowledging a multi-vendor future. They don't need to build every agent you use; they just want to be the layer that authenticates, manages, and secures them. It’s a bid to become the operating system for the agentic economy, ensuring that even if you use a competitor's model, you’re likely paying Microsoft to keep it in line.

🔍 The Rundown

📈 Trending Now: The End of Model Exclusivity?

For the last few years, the GenAI cloud wars were fought on the premise that "the best cloud has the best exclusive model." Microsoft had OpenAI; Google had DeepMind/Gemini; AWS leaned on Anthropic. This week, that dynamic fractured.

At Ignite, Microsoft announced that Anthropic’s Claude models are coming to Microsoft Foundry. This is an important message. Microsoft is effectively saying that Azure is a "Model Agnostic" platform. They are willing to host their biggest partner’s strongest rival (Claude) if it keeps developers within the Azure ecosystem. It signals that Microsoft believes the platform (tooling, governance, security) is a stronger moat than the model. 

Contrast this with Google’s approach. This week they doubled down on vertical integration with Gemini 3, boasting top-tier reasoning benchmarks and deep integration into the Workspace stack. Google is betting that superior, first-party models will still drive adoption. 

The Analysis: We are seeing a divergence in philosophy. Microsoft is becoming the "Windows" of AI—run whatever you want, as long as it's on our OS. Google is playing the "Apple" card—integrated, proprietary, and theoretically seamless. The risk for Google? If models become commodities and GPT-5, Claude 4, and Gemini 3 roughly converge on performance, the platform with the most flexibility (Azure/AWS) usually wins over the walled garden.

📅 Event Radar

👋 Until Next Week

Between the record-breaking 15 Tbps attack on Azure and the massive Cloudflare outage that took down ChatGPT and Bluesky due to a simple config error, this week was a reminder of the fragility of our digital backbone. As we rush toward "Agentic AI" where software autonomously spins up infrastructure, the complexity of managing these systems is about to skyrocket.

Also, keep an eye on the EU in the coming weeks—they have formally launched an investigation into whether AWS and Microsoft should be designated as "Gatekeepers," which could force radical changes to egress fees and licensing bundles.

More exciting news to come!

Do you enjoy these emails? Your friends and colleagues might, too! Help us grow the cloud community by sharing the newsletter with others.