<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Microsoft · AI Brief · darius.wiki</title><link>https://www.darius.wiki/en/topics/microsoft/</link><atom:link href="https://www.darius.wiki/en/topics/microsoft/rss.xml" rel="self" type="application/rss+xml"/><description>Microsoft's own models, Copilot and enterprise agent platform.</description><language>en</language><lastBuildDate>Sat, 21 Nov 2026 00:00:00 GMT</lastBuildDate><item><title>Microsoft Foundry Managed Compute is generally available: deploy 150+ open models on dedicated GPUs, choosing NVIDIA A100, H100, H200 or AMD MI300X</title><link>https://www.darius.wiki/en/daily/2026-11-21/foundry-managed-compute-ga/</link><guid isPermaLink="true">https://www.darius.wiki/en/daily/2026-11-21/foundry-managed-compute-ga/</guid><pubDate>Sat, 21 Nov 2026 00:00:00 GMT</pubDate><category>Industry</category><description>Hosting open models yourself means running GPUs, containers and the serving runtime. What is left for you to decide once Microsoft packages all that as a managed service. Take: Renting dedicated GPUs by the hour suits steady traffic and data-residency requirements; for spiky traffic, per-token managed APIs are usually cheaper. I'd work out daily token volume first and migrate only past the break-even point, and try an MI300X template while at it: a second GPU supplier is leverage in itself.</description></item><item><title>Microsoft Foundry lets custom agents have their own Microsoft 365 accounts: their own email, calendar, OneDrive and Teams identity, with a manager accountable for them</title><link>https://www.darius.wiki/en/daily/2026-11-20/foundry-agents-own-identity/</link><guid isPermaLink="true">https://www.darius.wiki/en/daily/2026-11-20/foundry-agents-own-identity/</guid><pubDate>Fri, 20 Nov 2026 00:00:00 GMT</pubDate><category>Industry</category><description>Most agents act with a user's borrowed credentials and permissions, which breaks down in group chats where it's unclear who they act for. Microsoft's identity answer, for teams building agents for enterprise collaboration. Take: The most practical item I read this week: the hard part of agents is not capability but who answers when something goes wrong. Give an agent its own account and an accountable manager, and audit, permissions and offboarding reuse the processes you already have for staff. If you're building an enterprise product, separate agent identity from acting on behalf of a user in your data model now, before a customer's security team asks.</description></item><item><title>Foundry Agent Service at Ignite: the managed Copilot harness enters public preview, while voice agents, computer use and browser automation become generally available</title><link>https://www.darius.wiki/en/daily/2026-11-19/foundry-agent-service-ignite/</link><guid isPermaLink="true">https://www.darius.wiki/en/daily/2026-11-19/foundry-agent-service-ignite/</guid><pubDate>Thu, 19 Nov 2026 00:00:00 GMT</pubDate><category>Tools</category><description>Microsoft now runs the agent loop and sandbox as a managed service. The main new capabilities across build, deploy and operate, for teams building enterprise agents on Azure. Take: Alongside Anthropic's self-hosted Managed Agents on 11-12 and AgentCore's new runtime, all three now sell the same thing: you write what the agent knows, and the platform owns the loop, sandbox and egress controls. My advice: start with an egress allowlist, which should be a default on any platform; the harness is still in preview, so do not lock production work to it yet.</description></item><item><title>Microsoft Ignite: MAI Code 2 and Code 2 Flash coding models, MAI Licensed Weights for trusted customers, and over 50 million agents in Agent 365</title><link>https://www.darius.wiki/en/daily/2026-11-18/microsoft-ignite-mai-code-2/</link><guid isPermaLink="true">https://www.darius.wiki/en/daily/2026-11-18/microsoft-ignite-mai-code-2/</guid><pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate><category>Model release</category><description>Nadella's Ignite keynote puts Microsoft's own models up front: the coding models went into GitHub Copilot the same day, and some customers can get the weights. Models, weight licensing and agent governance, for teams building enterprise products on Microsoft's stack. Take: MAI Code 2 is in Copilot's model picker, so Copilot teams can compare it on the same tasks today without wiring up another API. I would put Code 2 Flash on PandaClaws' routine back-end fixes for a week and judge it on rework rate and cost per task. The weight licensing is far from a team our size, but for large customers with data residency needs it is a new option worth noting.</description></item><item><title>GitHub and Microsoft scan six open source projects with AI: thousands of possible issues become 99 private reports, and 55 were fixed by November 15</title><link>https://www.darius.wiki/en/daily/2026-11-18/github-mdash-oss-fixes/</link><guid isPermaLink="true">https://www.darius.wiki/en/daily/2026-11-18/github-mdash-oss-fixes/</guid><pubDate>Wed, 18 Nov 2026 00:00:00 GMT</pubDate><category>Safety</category><description>AI makes finding vulnerabilities fast and cheap; the hard part is getting them fixed. How one pilot moved from dumping findings on maintainers to sending only verified reports with suggested fixes, for people building security scanners or maintaining open source. Take: Thousands down to 99 is the lesson in itself: the value of AI scanning is in filtering and verifying, not volume. When we wire up AI security scanning, I would require every result to come with reproduction steps and a patch, or not come at all. FastMCP is on the list, so teams building MCP servers on it should check recent release notes and make sure they have upgraded.</description></item></channel></rss>