<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Confidence score – M365.FM</title><link>https://www.m365.fm/entity/confidence-score/</link><description>An estimate attached to an individual AI extraction result that indicates how sure the system is about that value; it is not a guarantee that the value is correct.</description><language>en</language><item><title>How to Handle Low Confidence Scores in Document AI and Power Platform</title><link>https://www.m365.fm/community/u/mirko-p/blog/handling-low-confidence-scores-document-ai-power-platform/</link><guid>https://www.m365.fm/community/u/mirko-p/blog/handling-low-confidence-scores-document-ai-power-platform/</guid><pubDate>Fri, 18 Sep 2026 10:37:15 -0000</pubDate><category>article</category><description>Learn how to manage low confidence scores in Document AI and Power Platform with validation rules, Dataverse tracking, and Power Apps exception reviews.</description></item><item><title>Scaling Microsoft Copilot to 35 Million Pages: Lessons from the Epstein Files</title><link>https://www.m365.fm/community/u/mirko-p/blog/scaling-microsoft-copilot-to-35-million-pages-lessons-from-the-epstein-files/</link><guid>https://www.m365.fm/community/u/mirko-p/blog/scaling-microsoft-copilot-to-35-million-pages-lessons-from-the-epstein-files/</guid><pubDate>Wed, 26 Aug 2026 16:14:20 -0000</pubDate><category>article</category><description>Welcome back to the podcast blog! Today, we are expanding on one of our most fascinating episodes yet. When data scales into the millions, traditional tools inevita…</description></item><item><title>Scaling Microsoft Copilot for Massive Datasets: Lessons from 3.5 Million Pages</title><link>https://www.m365.fm/community/u/mirko-p/blog/scaling-microsoft-copilot-for-massive-datasets-lessons-from-35-million-pages/</link><guid>https://www.m365.fm/community/u/mirko-p/blog/scaling-microsoft-copilot-for-massive-datasets-lessons-from-35-million-pages/</guid><pubDate>Wed, 26 Aug 2026 09:52:45 -0000</pubDate><category>article</category><description>When you are tasked with processing massive amounts of unstructured data, traditional search tools quickly hit a brick wall. In modern enterprise environments and c…</description></item></channel></rss>