<?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>Red-Teaming on Ashwin Viswamithiran</title><link>https://ashwinviswamithiran.com/tags/red-teaming/</link><description>Recent content in Red-Teaming on Ashwin Viswamithiran</description><image><title>Ashwin Viswamithiran</title><url>https://ashwinviswamithiran.com/og-default.png</url><link>https://ashwinviswamithiran.com/og-default.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 11 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://ashwinviswamithiran.com/tags/red-teaming/index.xml" rel="self" type="application/rss+xml"/><item><title>Twelve Ways to Make a Threat-Intel Bot Lie</title><link>https://ashwinviswamithiran.com/writing/twelve-ways-to-make-a-threat-intel-bot-lie/</link><pubDate>Thu, 11 Jun 2026 00:00:00 +0000</pubDate><guid>https://ashwinviswamithiran.com/writing/twelve-ways-to-make-a-threat-intel-bot-lie/</guid><description>An evaluation of a retrieval-grounded threat-intelligence assistant against twelve adversarial prompts designed to induce fabrication, a twenty-four query quality set, and a three-model comparison — including what the tests failed to cover.</description></item></channel></rss>