<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Threat Modeling on Willis Vandevanter</title><link>https://silentrobots.com/tags/threat-modeling/</link><description>Recent content in Threat Modeling on Willis Vandevanter</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Thu, 06 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://silentrobots.com/tags/threat-modeling/index.xml" rel="self" type="application/rss+xml"/><item><title>rethinking threat models when the cost is tokens not time</title><link>https://silentrobots.com/rethinking-threat-models-when-the-cost-is-tokens-not-time/</link><pubDate>Thu, 06 Aug 2026 00:00:00 +0000</pubDate><guid>https://silentrobots.com/rethinking-threat-models-when-the-cost-is-tokens-not-time/</guid><description>&lt;img src="https://silentrobots.com/" alt="Featured image of post rethinking threat models when the cost is tokens not time" /&gt;&lt;p&gt;another by-product of &lt;a class="link" href="https://huggingface.co/blog/agent-intrusion-technical-timeline" target="_blank" rel="noopener"
 &gt;anatomy of a frontier lab agent intrusion: a technical timeline of the july 2026 incident&lt;/a&gt; is that the adversary mindset is not the same. &lt;a class="link" href="https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing" target="_blank" rel="noopener"
 &gt;puppet accounts&lt;/a&gt;, &lt;a class="link" href="https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing" target="_blank" rel="noopener"
 &gt;social engineering&lt;/a&gt; and &lt;a class="link" href="https://huggingface.co/blog/agent-intrusion-technical-timeline" target="_blank" rel="noopener"
 &gt;grinding through k8s manifests&lt;/a&gt; were all part of the process for the agents. one caveat is these reports are not based around token costs as they were frontier labs. but in the not so distant future agent exploitation will have a new calculation in token cost. we already see this in cyber products. similarly the trade off from agent to human will have cost engineering front and center. but fwiw agent goals are not necessarily the same as the operator goals. appeasing the operator and agents achieving goal regardless of cost are two distinct vectors.&lt;/p&gt;</description></item></channel></rss>