<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>The AI Lyceum — Newsletter by Samraj Matharu</title>
    <link>https://advertisingprotocols.com/newsletter</link>
    <description>Weekly insights on AI, advertising protocols, philosophy of intelligence, and agentic systems. Wisdom over Intelligence.</description>
    <language>en</language>
    <lastBuildDate>Mon, 29 Sep 2026 09:00:00 GMT</lastBuildDate>
    <atom:link href="https://advertisingprotocols.com/newsletter.xml" rel="self" type="application/rss+xml"/>
    <image>
      <url>https://advertisingprotocols.com/og-image.png</url>
      <title>The AI Lyceum</title>
      <link>https://advertisingprotocols.com/newsletter</link>
    </image>

    <item>
      <title>Before Artificial Intelligence, We Need Decision Intelligence</title>
      <link>https://www.linkedin.com/pulse/before-artificial-intelligence-we-need-decision-samraj-matharu--l5mbe</link>
      <description>Organisations should map, simplify and test their decision processes before adopting AI tools. Decision intelligence makes explicit how information and judgment lead to outcomes — then systematically improves them.</description>
      <category>Decision Intelligence</category>
      <pubDate>Mon, 29 Sep 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="false">newsletter-2026-09-29-decision-intelligence</guid>
    </item>
    <item>
      <title>What is Jev? The AI model built for decisions</title>
      <link>https://www.linkedin.com/pulse/what-jev-ai-model-built-decisions-samraj-matharu--v9jae</link>
      <description>Introducing Jev — a new kind of AI model designed not to generate text, but to make structured decisions. How TypeSafe's System One models turn natural language into typed judgments and probabilities.</description>
      <category>AI</category>
      <pubDate>Sat, 27 Sep 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="false">newsletter-2026-09-27-jev</guid>
    </item>
    <item>
      <title>The Man Who Formalised Reasoning, 2,300 Years Before AI.</title>
      <link>https://www.linkedin.com/pulse/man-who-formalised-reasoning-2300-years-before-ai-samraj-matharu--zebye</link>
      <description>Aristotle's syllogistic logic laid the groundwork for every reasoning system we use today — from formal proofs to large language models. A look at how ancient philosophy still shapes modern AI.</description>
      <category>Philosophy</category>
      <pubDate>Fri, 26 Sep 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="false">newsletter-2026-09-26-aristotle</guid>
    </item>
    <item>
      <title>Autotelic AI: When AI sets its own goals, who decides what matters?</title>
      <link>https://www.linkedin.com/pulse/autotelic-ai-when-sets-its-own-goals-who-decides-what-samraj-matharu--glkie</link>
      <description>Autotelic systems pursue self-generated goals without external reward signals. What happens when AI moves beyond instruction-following to intrinsic motivation — and what it means for alignment.</description>
      <category>AI Safety</category>
      <pubDate>Thu, 25 Sep 2026 09:00:00 GMT</pubDate>
      <guid isPermaLink="false">newsletter-2026-09-25-autotelic</guid>
    </item>
  </channel>
</rss>
