<?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/8ffd263ccb504ae08e614d46c8d16264&quot; frameborder=&quot;0&quot; width=&quot;1348&quot; height=&quot;1011&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1011</height><width>1348</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1011</thumbnail_height><thumbnail_width>1348</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/8ffd263ccb504ae08e614d46c8d16264-5c5300c10fd1558c.gif</thumbnail_url><duration>296.368</duration><title>Introducing Policy AI</title><description>In this video, I’m excited to introduce Policy AI, a powerful content moderation platform that I’ve always wanted during my time at Grindr and OkCupid. This tool leverages state-of-the-art LLMs to enforce custom policies tailored to specific platforms, allowing for nuanced moderation of various content types, including text, images, and videos. I demonstrated how to run policies in parallel and highlighted our impressive F1 score of 100% after a quick adjustment to address spam. I encourage you to explore how Policy AI can enhance your content moderation strategies. Please let me know if you have any questions or need further assistance!</description></oembed>