<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/be929480022148698f2d7d85ac143e56&quot; frameborder=&quot;0&quot; width=&quot;1788&quot; height=&quot;1341&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1341</height><width>1788</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1341</thumbnail_height><thumbnail_width>1788</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/be929480022148698f2d7d85ac143e56-6a340347517d8a33.gif</thumbnail_url><duration>256.6971</duration><title>AI Powered Page Classification with the Compliance API 🚀</title><description>In this video, I demonstrate an interesting use case of the new compliance API for Confluence, specifically focusing on automating page classification based on content. I walk through how I set up an automation that utilizes the get available levels for a page and set page classification endpoints, along with a custom Rover agent to recommend classification levels. I showcase two examples: an internal employee policy classified as &quot;internal&quot; and a company cardholders page classified as &quot;highly restricted.&quot; I also mention that future improvements could include factoring in detection information for more robust classifications. I encourage viewers to consider how they might implement similar automation in their own projects.

Anyone with access to the Confluence demo instance can test this new automation out by creating a new page in Project Alpha</description></oembed>