<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Double Agent AI on BriefArc</title><link>https://briefarc.com/tags/double-agent-ai/</link><description>Recent content in Double Agent AI on BriefArc</description><image><title>BriefArc</title><url>https://briefarc.com/images/og-cover.png</url><link>https://briefarc.com/images/og-cover.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 11 May 2026 14:31:05 +0000</lastBuildDate><atom:link href="https://briefarc.com/tags/double-agent-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>New Tool Spots When Your AI Assistant Is Secretly Working Against You</title><link>https://briefarc.com/posts/new-tool-spots-when-your-ai-assistant-is-secretly-working-against-you/</link><pubDate>Mon, 11 May 2026 14:31:05 +0000</pubDate><guid>https://briefarc.com/posts/new-tool-spots-when-your-ai-assistant-is-secretly-working-against-you/</guid><description>A team at Rochester Institute of Technology has developed a tool that can detect when an AI agent has been compromised—acting as a &amp;#39;double agent&amp;#39;—and what you can do to stay safe.</description></item></channel></rss>