<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Scott Sloan on Medium]]></title>
        <description><![CDATA[Stories by Scott Sloan on Medium]]></description>
        <link>https://medium.com/@swsloan?source=rss-a4a753ac8f49------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*x3Qw3SyI2GN899v-Atfxlg.jpeg</url>
            <title>Stories by Scott Sloan on Medium</title>
            <link>https://medium.com/@swsloan?source=rss-a4a753ac8f49------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:44:29 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@swsloan/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Three Rules That Don’t Bend: Keeping an AI SOC Defensible as the Models Get Better]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/three-rules-that-dont-bend-keeping-an-ai-soc-defensible-as-the-models-get-better-566445137c47?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*NulsINR1X_PtidbX.png" width="1200"></a></p><p class="medium-feed-snippet">Models will keep improving. These three principles shouldn&#x2019;t change when they do.</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/three-rules-that-dont-bend-keeping-an-ai-soc-defensible-as-the-models-get-better-566445137c47?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/three-rules-that-dont-bend-keeping-an-ai-soc-defensible-as-the-models-get-better-566445137c47?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/566445137c47</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Sun, 25 Oct 2026 14:01:03 GMT</pubDate>
            <atom:updated>2026-10-25T14:01:03.578Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The First Ninety Days of an AI Agentic SOC: A Week-by-Week Plan]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/the-first-ninety-days-of-an-ai-agentic-soc-a-week-by-week-plan-13e3a7e865b1?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*Wj3STpk2yu_HjQG5.png" width="1200"></a></p><p class="medium-feed-snippet">Three thirty-day blocks, each with a clear definition of done, turning the whole playbook into a schedule you can put in front of your&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/the-first-ninety-days-of-an-ai-agentic-soc-a-week-by-week-plan-13e3a7e865b1?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/the-first-ninety-days-of-an-ai-agentic-soc-a-week-by-week-plan-13e3a7e865b1?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/13e3a7e865b1</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[llm]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Fri, 23 Oct 2026 14:01:04 GMT</pubDate>
            <atom:updated>2026-10-23T14:01:04.057Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Climb the Rollout Ladder: Taking an AI Triage Agent From Shadow Mode to Steady State]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/climb-the-rollout-ladder-taking-an-ai-triage-agent-from-shadow-mode-to-steady-state-e04a90a8868e?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*YBSvLMVsOseNY5AN.png" width="1200"></a></p><p class="medium-feed-snippet">Four rungs, each with entry criteria, warning signs and a way back down. Don&#x2019;t skip one, and be willing to stay on a rung for months.</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/climb-the-rollout-ladder-taking-an-ai-triage-agent-from-shadow-mode-to-steady-state-e04a90a8868e?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/climb-the-rollout-ladder-taking-an-ai-triage-agent-from-shadow-mode-to-steady-state-e04a90a8868e?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/e04a90a8868e</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Wed, 21 Oct 2026 14:01:03 GMT</pubDate>
            <atom:updated>2026-10-21T14:01:03.359Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Measure the Two Errors Separately: The Metrics That Matter for an AI Triage Agent]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/measure-the-two-errors-separately-the-metrics-that-matter-for-an-ai-triage-agent-206af5e1f2be?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*KKjnU02IDBGj-MWg.png" width="1200"></a></p><p class="medium-feed-snippet">Accuracy is the wrong number. Here are the six measures that tell you whether an agent is safe to rely on, and how to report them honestly.</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/measure-the-two-errors-separately-the-metrics-that-matter-for-an-ai-triage-agent-206af5e1f2be?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/measure-the-two-errors-separately-the-metrics-that-matter-for-an-ai-triage-agent-206af5e1f2be?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/206af5e1f2be</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Mon, 19 Oct 2026 14:01:09 GMT</pubDate>
            <atom:updated>2026-10-19T14:01:09.161Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Making an AI Triage Agent Auditable: Frameworks to Map To, and What to Log]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/making-an-ai-triage-agent-auditable-frameworks-to-map-to-and-what-to-log-174443b1486b?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*4IdszYuxbwzC7P5g.png" width="1200"></a></p><p class="medium-feed-snippet">Your auditors will ask how the agent is governed and why it made a particular call. Here&#x2019;s how to have both answers ready before they ask.</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/making-an-ai-triage-agent-auditable-frameworks-to-map-to-and-what-to-log-174443b1486b?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/making-an-ai-triage-agent-auditable-frameworks-to-map-to-and-what-to-log-174443b1486b?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/174443b1486b</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Sat, 17 Oct 2026 14:01:02 GMT</pubDate>
            <atom:updated>2026-10-17T14:01:02.068Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Break the Lethal Trifecta: The One Capability Your AI Triage Agent Must Never Have]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/break-the-lethal-trifecta-the-one-capability-your-ai-triage-agent-must-never-have-cdcb107f4325?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*B_H00QdC33zcoSgx.png" width="1200"></a></p><p class="medium-feed-snippet">A SOC agent inherently holds private data and reads untrusted content. The whole security design comes down to whether it can also talk to&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/break-the-lethal-trifecta-the-one-capability-your-ai-triage-agent-must-never-have-cdcb107f4325?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/break-the-lethal-trifecta-the-one-capability-your-ai-triage-agent-must-never-have-cdcb107f4325?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/cdcb107f4325</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Thu, 15 Oct 2026 14:01:04 GMT</pubDate>
            <atom:updated>2026-10-15T14:01:04.400Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Prompt Injection Hides in Your Alert Data: Where It Arrives, and Six Defenses Ranked]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/prompt-injection-hides-in-your-alert-data-where-it-arrives-and-six-defenses-ranked-17730fa3322c?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*PjQbRasX6WsJ4kJt.png" width="1200"></a></p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/prompt-injection-hides-in-your-alert-data-where-it-arrives-and-six-defenses-ranked-17730fa3322c?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/prompt-injection-hides-in-your-alert-data-where-it-arrives-and-six-defenses-ranked-17730fa3322c?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/17730fa3322c</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Tue, 13 Oct 2026 14:01:09 GMT</pubDate>
            <atom:updated>2026-10-13T14:01:09.255Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[The AI Triage Pipeline: Why the Model Never Acts Alone]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/the-ai-triage-pipeline-why-the-model-never-acts-alone-06a48cd9ed00?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*bdtUTb3mBDAGgNMq.png" width="1200"></a></p><p class="medium-feed-snippet">A seven-step architecture for an alert triage agent where plain code does most of the work, and the model does only the part a model is&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/the-ai-triage-pipeline-why-the-model-never-acts-alone-06a48cd9ed00?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/the-ai-triage-pipeline-why-the-model-never-acts-alone-06a48cd9ed00?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/06a48cd9ed00</guid>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Sun, 11 Oct 2026 14:01:02 GMT</pubDate>
            <atom:updated>2026-10-11T14:01:02.897Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[No Golden Set? Start Here: Documenting SOC Cases So Auditors, Analysts and AI Can Learn From Them]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/no-golden-set-start-here-documenting-soc-cases-so-auditors-analysts-and-ai-can-learn-from-them-ee8c197d55db?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*Y3i8ugTnNsolg9Tg.png" width="1200"></a></p><p class="medium-feed-snippet">Most SOCs don&#x2019;t have 300 well-documented closed cases to test an AI triage agent against. Here&#x2019;s why that gap matters well beyond AI, what&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/no-golden-set-start-here-documenting-soc-cases-so-auditors-analysts-and-ai-can-learn-from-them-ee8c197d55db?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/no-golden-set-start-here-documenting-soc-cases-so-auditors-analysts-and-ai-can-learn-from-them-ee8c197d55db?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/ee8c197d55db</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 14:01:05 GMT</pubDate>
            <atom:updated>2026-10-09T14:01:05.298Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Measure Before You Build: Golden Sets, Evaluation Harnesses and Model Screening for an AI SOC]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@swsloan/measure-before-you-build-golden-sets-evaluation-harnesses-and-model-screening-for-an-ai-soc-5f6656f60e59?source=rss-a4a753ac8f49------2"><img src="https://cdn-images-1.medium.com/max/1200/0*CwB8bwCuUZWktawa.png" width="1200"></a></p><p class="medium-feed-snippet">The first thirty days of an AI SOC project shouldn&#x2019;t produce an agent. They should produce the ability to tell whether any agent is any&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@swsloan/measure-before-you-build-golden-sets-evaluation-harnesses-and-model-screening-for-an-ai-soc-5f6656f60e59?source=rss-a4a753ac8f49------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@swsloan/measure-before-you-build-golden-sets-evaluation-harnesses-and-model-screening-for-an-ai-soc-5f6656f60e59?source=rss-a4a753ac8f49------2</link>
            <guid isPermaLink="false">https://medium.com/p/5f6656f60e59</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[cybersecurity]]></category>
            <dc:creator><![CDATA[Scott Sloan]]></dc:creator>
            <pubDate>Wed, 07 Oct 2026 14:01:06 GMT</pubDate>
            <atom:updated>2026-10-07T14:01:06.224Z</atom:updated>
        </item>
    </channel>
</rss>