<?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/e846c2820dd84c77b24250234590de2e&quot; frameborder=&quot;0&quot; width=&quot;1108&quot; height=&quot;831&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>831</height><width>1108</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>831</thumbnail_height><thumbnail_width>1108</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/e846c2820dd84c77b24250234590de2e-8944fea611465af1.gif</thumbnail_url><duration>219.789</duration><title>Loom Canvas - 27 September 2026</title><description>This Loom explains how the Sophron AI interface was built to reduce friction and improve user privacy during voice-based conversations. The speaker says they often experienced the AI talking over them when they paused, and they were concerned that their data was being sent to another server and stored in that provider’s databases. Sophron uses a ChatGPT-like interface with a microphone and text, running on Quinn models trained to ask Socratic questions, and it can reference local notes via a vault inspired by Obsidian. The presenter emphasizes that the AI is designed to guide users with questions rather than provide direct answers, helping users arrive at their own insights.</description></oembed>