<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>The Microsoft Cloud Blog</title>
    <link>https://themicrosoftcloudblog.com</link>
    <description>Insights and expertise on Microsoft Azure, Cloud Computing, and Enterprise Technology</description>
    <language>en-us</language>
    <lastBuildDate>Sat, 25 Apr 2026 10:46:08 GMT</lastBuildDate>
    <atom:link href="https://themicrosoftcloudblog.com/feed.xml" rel="self" type="application/rss+xml"/>
    <image>
      <url>https://themicrosoftcloudblog.com/images/richard-hogan.jpeg</url>
      <title>The Microsoft Cloud Blog</title>
      <link>https://themicrosoftcloudblog.com</link>
    </image>
    
    <item>
      <title><![CDATA[What Your Agent Remembers, and Who Controls It]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/what-your-agent-remembers-and-who-controls-it/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/what-your-agent-remembers-and-who-controls-it/</guid>
      <pubDate>Fri, 24 Apr 2026 10:36:43 GMT</pubDate>
      <description><![CDATA[As AI agents move from experimental to operational, the question of what they remember, how accurately they remember it, and who controls that memory is becoming one of the more important architectural decisions organisations will make. It is also one of the least examined. This piece maps the territory, from the trade-offs between different memory architectures to the compliance, governance, and security implications that follow.]]></description>
      <category><![CDATA[Reaching for the Cloud]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Microsoft Discovery: What the Platform Actually Does Beneath the Announcement]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/microsoft-discovery-what-the-platform-actually-does-beneath-the-announcement/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/microsoft-discovery-what-the-platform-actually-does-beneath-the-announcement/</guid>
      <pubDate>Thu, 23 Apr 2026 10:06:43 GMT</pubDate>
      <description><![CDATA[Microsoft expanded preview access to Microsoft Discovery this week, but the announcement leads with case studies rather than architecture. The platform's actual design (a graph-based knowledge foundation, a reasoning-first Discovery Engine, and agentic execution across HPC and physical lab infrastructure) is what distinguishes it from general-purpose AI tools, and it is worth understanding before the hype settles.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Copilot]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[GitHub Copilot's Pricing Problem Is an Agentic Problem]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/github-copilot-s-pricing-problem-is-an-agentic-problem/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/github-copilot-s-pricing-problem-is-an-agentic-problem/</guid>
      <pubDate>Tue, 21 Apr 2026 10:09:26 GMT</pubDate>
      <description><![CDATA[GitHub is pausing new Copilot sign-ups and tightening usage limits, and the reason it gives is more interesting than the changes themselves. Agentic workflows have broken the economics of flat-rate AI tooling, and GitHub is the first major vendor to say so publicly.]]></description>
      <category><![CDATA[GitHub]]></category><category><![CDATA[Copilot]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[GitHub Copilot Modernisation Is What AI Agents Should Actually Look Like]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/github-copilot-modernisation-is-what-ai-agents-should-actually-look-like/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/github-copilot-modernisation-is-what-ai-agents-should-actually-look-like/</guid>
      <pubDate>Fri, 17 Apr 2026 10:48:26 GMT</pubDate>
      <description><![CDATA[AI agents are most useful when they are honest about their limits. GitHub Copilot Modernisation does the mechanical work of .NET migration well, and its assess-plan-execute structure is a better model for agentic design than most of what gets discussed in the abstract.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[GitHub]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[The Productivity Mirage: Why AI Is Changing Work Without Necessarily Improving It]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/the-productivity-mirage-why-ai-is-changing-work-without-necessarily-improving-it/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/the-productivity-mirage-why-ai-is-changing-work-without-necessarily-improving-it/</guid>
      <pubDate>Fri, 10 Apr 2026 10:24:27 GMT</pubDate>
      <description><![CDATA[Microsoft Research's annual New Future of Work report arrived this week carrying a finding that deserves more attention than it will probably get: AI is not automatically making organisations more productive, and the people most at risk from it are the ones organisations can least afford to lose.]]></description>
      <category><![CDATA[Reaching for the Cloud]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[The AI Bill Arrived. Now Everyone Wants to Talk About ROI.]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/the-ai-bill-arrived-now-everyone-wants-to-talk-about-roi/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/the-ai-bill-arrived-now-everyone-wants-to-talk-about-roi/</guid>
      <pubDate>Thu, 09 Apr 2026 06:31:18 GMT</pubDate>
      <description><![CDATA[Microsoft's new Cloud Cost Optimization series is a tacit acknowledgement that the early wave of AI investment did not come with adequate accountability attached. The real problem is governance — specifically, the gap between having a governance framework and having the operational visibility that actually makes ROI measurable.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Governance]]></category><category><![CDATA[ROI]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[The Governance Gap: Why AI Agents Are Exposing an Older Problem]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/the-governance-gap-why-ai-agents-are-exposing-an-older-problem/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/the-governance-gap-why-ai-agents-are-exposing-an-older-problem/</guid>
      <pubDate>Sun, 05 Apr 2026 09:00:00 GMT</pubDate>
      <description><![CDATA[Most organisations are not struggling with AI agents because the technology is unsafe. They are struggling because their governance models were built for a world that no longer exists. The problem is older than the agents, and the answers are harder than most frameworks are currently prepared to admit.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Governance]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Your Agent Estate Is the Next Attack Surface — and Most Organisations Can't Even Inventory It]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/your-agent-estate-is-the-next-attack-surface-and-most-organisations-can-t-even-inventory-it/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/your-agent-estate-is-the-next-attack-surface-and-most-organisations-can-t-even-inventory-it/</guid>
      <pubDate>Sat, 04 Apr 2026 09:00:00 GMT</pubDate>
      <description><![CDATA[AI is no longer just helping threat actors work faster. It is embedded across the entire attack lifecycle, and the next major target is the agent ecosystem most organisations are deploying without mature identity, inventory, or governance controls. Microsoft's RSAC 2026 briefing lays out why the agent surface problem is the one architects need to prioritise now.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Copilot]]></category><category><![CDATA[Governance]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Multi-Agent Orchestration Goes GA: What the Latest Copilot Studio Update Means for Enterprise Architects]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/multi-agent-orchestration-goes-ga-what-the-latest-copilot-studio-update-means-for-enterprise-architects/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/multi-agent-orchestration-goes-ga-what-the-latest-copilot-studio-update-means-for-enterprise-architects/</guid>
      <pubDate>Fri, 03 Apr 2026 09:00:00 GMT</pubDate>
      <description><![CDATA[Microsoft's March 2026 Copilot Studio update brings multi-agent orchestration to general availability, and for enterprise architects, it changes the default design assumption. The question is no longer whether your agents can coordinate across teams and platforms. It is whether you have designed them to.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Copilot Studio]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Why Your Existing Tools Probably Can't Solve the AI Governance Problem]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/why-your-existing-tools-probably-can-t-solve-the-ai-governance-problem/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/why-your-existing-tools-probably-can-t-solve-the-ai-governance-problem/</guid>
      <pubDate>Thu, 02 Apr 2026 12:00:00 GMT</pubDate>
      <description><![CDATA[Your organisation has diagramming tools, a CMDB, an architecture repository, and a risk register. So why can't any of them tell you what your AI systems are actually authorised to do? The tools aren't broken. They were just never built to hold this kind of information.]]></description>
      <category><![CDATA[Reaching for the Cloud]]></category><category><![CDATA[Architecture]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Sovereign AI at the Edge: It's Not Just About Where the Data Lives]]></title>
      <link>https://themicrosoftcloudblog.com/2026/04/sovereign-ai-at-the-edge-it-s-not-just-about-where-the-data-lives/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/04/sovereign-ai-at-the-edge-it-s-not-just-about-where-the-data-lives/</guid>
      <pubDate>Wed, 01 Apr 2026 13:10:58 GMT</pubDate>
      <description><![CDATA[Microsoft and Armada's collaboration to deliver Azure Local on Galleon modular datacenters is more than an edge infrastructure announcement. It reframes what sovereign cloud actually means. It's not just about where data lives. It's about where intelligence runs, and who controls it.]]></description>
      <category><![CDATA[Azure]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Agentic AI and the OWASP Top 10: What Security Teams Need to Do Now]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/agentic-ai-and-the-owasp-top-10-what-security-teams-need-to-do-now/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/agentic-ai-and-the-owasp-top-10-what-security-teams-need-to-do-now/</guid>
      <pubDate>Tue, 31 Mar 2026 08:56:24 GMT</pubDate>
      <description><![CDATA[The OWASP Top 10 for Agentic Applications (2026) is out, and Microsoft has mapped it to Copilot Studio and the upcoming Agent 365 platform. Here is what security and architecture teams need to take from it.]]></description>
      <category><![CDATA[Cybersecurity]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Not All Assets Are Equal: How Microsoft Defender Is Rethinking Protection for High-Value Systems]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/not-all-assets-are-equal-how-microsoft-defender-is-rethinking-protection-for-high-value-systems/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/not-all-assets-are-equal-how-microsoft-defender-is-rethinking-protection-for-high-value-systems/</guid>
      <pubDate>Mon, 30 Mar 2026 07:12:28 GMT</pubDate>
      <description><![CDATA[Microsoft Defender's high-value asset protection framework applies differentiated detection and prevention logic based on a system's role in the environment — meaning a domain controller gets stronger protection than a general workstation even when the same activity is observed on both. Here is how it works, and what you need to do to make it effective in your environment.]]></description>
      <category><![CDATA[Cybersecurity]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[GroundedPlanBench: Spatially grounded long-horizon task planning for robot manipulation]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/groundedplanbench-spatially-grounded-long-horizon-task-planning-for-robot-manipulation/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/groundedplanbench-spatially-grounded-long-horizon-task-planning-for-robot-manipulation/</guid>
      <pubDate>Thu, 26 Mar 2026 18:12:13 GMT</pubDate>
      <description><![CDATA[The announcement of GroundedPlanBench by Microsoft Research stands out in the ongoing evolution of robot manipulation, particularly for tasks that span multiple steps and require nuanced spatial awareness]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Robotics]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Your AI systems are running. Does anyone know what they're permitted to do?]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/nobody-knows-what-their-ai-systems-are-actually-allowed-to-do/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/nobody-knows-what-their-ai-systems-are-actually-allowed-to-do/</guid>
      <pubDate>Thu, 26 Mar 2026 11:03:03 GMT</pubDate>
      <description><![CDATA[Most organisations deploying AI can't answer a simple question: what is this system actually authorised to do? With the EU AI Act creating real pressure and Microsoft's own security researchers flagging the same gap, the documentation debt from two years of fast AI deployment is about to get expensive.]]></description>
      <category><![CDATA[Reaching for the Cloud]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Guidance for detecting, investigating, and defending against the Trivy supply chain compromise]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/guidance-for-detecting-investigating-and-defending-against-the-trivy-supply-chain-compromise/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/guidance-for-detecting-investigating-and-defending-against-the-trivy-supply-chain-compromise/</guid>
      <pubDate>Wed, 25 Mar 2026 13:16:20 GMT</pubDate>
      <description><![CDATA[The recent compromise of Trivy, Aqua Security’s well-regarded open-source vulnerability scanner, is a sobering moment for the security community]]></description>
      <category><![CDATA[Cybersecurity]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[What’s new with Microsoft in open-source and Kubernetes at KubeCon + CloudNativeCon Europe 2026]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/what-s-new-with-microsoft-in-open-source-and-kubernetes-at-kubecon-cloudnativecon-europe-2026/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/what-s-new-with-microsoft-in-open-source-and-kubernetes-at-kubecon-cloudnativecon-europe-2026/</guid>
      <pubDate>Tue, 24 Mar 2026 08:58:22 GMT</pubDate>
      <description><![CDATA[Microsoft’s announcements at KubeCon + CloudNativeCon Europe 2026 are a clear signal that the convergence of AI and Kubernetes is no longer theoretical—it is shaping the next phase of enterprise cloud infrastructure.]]></description>
      <category><![CDATA[Azure]]></category><category><![CDATA[Kubernetes]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Building Security for Agentic AI: A Strategic Perspective on End-to-End Trust]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/building-security-for-agentic-ai-a-strategic-perspective-on-end-to-end-trust/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/building-security-for-agentic-ai-a-strategic-perspective-on-end-to-end-trust/</guid>
      <pubDate>Mon, 23 Mar 2026 13:41:04 GMT</pubDate>
      <description><![CDATA[Agentic AI is no longer a futuristic concept—it’s rapidly becoming boardroom reality for CISOs and tech leaders alike. 🤖

At RSAC’s milestone 35th year, Microsoft spotlighted an eye-opening stat: 80% of Fortune 500 companies now deploy intelligent agents. The scale is impressive, but so are the security challenges that come with defending automated systems that act on their own.]]></description>
      <category><![CDATA[AI]]></category><category><![CDATA[Cybersecurity]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[Certifications Shifting, AI on the Perimeter, and the Real Meaning of Vibe Coding]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/certifications-shifting-ai-on-the-perimeter-and-the-real-meaning-of-vibe-coding/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/certifications-shifting-ai-on-the-perimeter-and-the-real-meaning-of-vibe-coding/</guid>
      <pubDate>Mon, 23 Mar 2026 09:27:51 GMT</pubDate>
      <description><![CDATA[Microsoft is reshaping certifications, identity is becoming the real perimeter, and AI is quietly changing what it even means to build an application. In this episode, a routine set of updates turns into a much bigger conversation about governance, judgement, and the realities of vibe coding in an AI first world.]]></description>
      <category><![CDATA[Azure]]></category><category><![CDATA[AI]]></category><category><![CDATA[Purview]]></category><category><![CDATA[Entra]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
    <item>
      <title><![CDATA[New tools and guidance: Announcing Zero Trust for AI]]></title>
      <link>https://themicrosoftcloudblog.com/2026/03/new-tools-and-guidance-announcing-zero-trust-for-ai/</link>
      <guid isPermaLink="true">https://themicrosoftcloudblog.com/2026/03/new-tools-and-guidance-announcing-zero-trust-for-ai/</guid>
      <pubDate>Fri, 20 Mar 2026 10:46:47 GMT</pubDate>
      <description><![CDATA[Microsoft’s announcement of Zero Trust for AI (ZT4AI) marks a significant moment in enterprise security thinkin]]></description>
      <category><![CDATA[Cybersecurity]]></category><category><![CDATA[Identity]]></category><category><![CDATA[Entra]]></category>
      <author>richard@themicrosoftcloudblog.com (Richard Hogan)</author>
    </item>
  </channel>
</rss>