Update Part 10 indexes and nav links for new frontier sections

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: apartsin

part-10-frontiers/module-34-emerging-architectures/index.html

@@ -104,6 +104,51 @@ <h2>Sections</h2>
                 </span>
             </a>
         </li>
+        <li>
+            <a href="section-34.5.html" class="section-card">
+                <span class="section-num">34.5</span>
+                <span class="section-title">A Theory of Reasoning in LLMs</span>
+                <span class="section-desc">
+                    Chain-of-thought as emergent computation, formal frameworks for reasoning, process reward models, compositional reasoning limits, and connections to cognitive science.
+                </span>
+            </a>
+        </li>
+        <li>
+            <a href="section-34.6.html" class="section-card">
+                <span class="section-num">34.6</span>
+                <span class="section-title">Memory as a Computational Primitive</span>
+                <span class="section-desc">
+                    Memory architectures beyond context windows (MemGPT, Letta), working memory vs. long-term memory, memory consolidation, and external memory as a Turing-completeness enabler.
+                </span>
+            </a>
+        </li>
+        <li>
+            <a href="section-34.7.html" class="section-card">
+                <span class="section-num">34.7</span>
+                <span class="section-title">Mechanistic Interpretability at Scale</span>
+                <span class="section-desc">
+                    Sparse autoencoders for feature discovery, circuit analysis, superposition and polysemanticity, scaling interpretability to frontier models, and practical applications for debugging and safety.
+                </span>
+            </a>
+        </li>
+        <li>
+            <a href="section-34.8.html" class="section-card">
+                <span class="section-num">34.8</span>
+                <span class="section-title">The Nature of Agency: When Does a Model Become an Agent?</span>
+                <span class="section-desc">
+                    Definitional frameworks for agency, degrees of autonomy, philosophical and engineering implications, and safety considerations including instrumental convergence and mesa-optimization.
+                </span>
+            </a>
+        </li>
+        <li>
+            <a href="section-34.9.html" class="section-card">
+                <span class="section-num">34.9</span>
+                <span class="section-title">Efficient Multi-Tool Orchestration and Tool Economy</span>
+                <span class="section-desc">
+                    Token-efficient tool calling patterns, tool routing and caching, parallel execution, economic models for tool use, and benchmarking tool efficiency.
+                </span>
+            </a>
+        </li>
     </ul>
 
 <div class="whats-next">

part-10-frontiers/module-34-emerging-architectures/section-34.4.html

@@ -884,9 +884,9 @@ <h2>7. Practical Lab: Video Prediction with a Diffusion Transformer <span class=
         <h2>What Comes Next</h2>
         <p>
             This section explored how world models extend the prediction paradigm from text to physical reality.
-            The next chapter, <a class="cross-ref" href="../module-35-ai-society/section-35.1.html">Chapter 35: AI and Society</a>,
-            shifts from technical architecture to the broader societal context in which these systems are deployed,
-            examining economic impact, governance, and the evolving relationship between humans and AI.
+            In the next section, <a class="cross-ref" href="section-34.5.html">Section 34.5: A Theory of Reasoning in LLMs</a>,
+            we examine formal frameworks for understanding when and why LLMs can reason, the role of chain-of-thought
+            as a computational primitive, and the limits of compositional reasoning.
         </p>
     </div>
 
@@ -910,7 +910,7 @@ <h2>Bibliography and Further Reading</h2>
 <nav class="chapter-nav">
     <a class="prev" href="section-34.3.html">Alternative Architectures Beyond Transformers</a>
     <a class="up" href="index.html">Chapter 34: Emerging Architectures &amp; Scaling Frontiers</a>
-    <a class="next" href="../module-35-ai-society/index.html">Chapter 35: AI, Society &amp; Open Problems</a>
+    <a class="next" href="section-34.5.html">A Theory of Reasoning in LLMs</a>
 </nav>
 
     <footer>

part-10-frontiers/module-35-ai-society/section-35.4.html

@@ -471,14 +471,7 @@ <h2>7. Building a Personal Research Agenda <span class="level-badge basic" title
 
 <div class="whats-next">
         <h2>What Comes Next</h2>
-        <p>
-            This is the final section of the book. You now have a comprehensive map of the field, from
-            foundational concepts in <a class="cross-ref" href="../../part-1-foundations/module-00-ml-pytorch-foundations/index.html">Part I</a>
-            through production engineering and safety, all the way to the open problems that will define
-            the next era of AI. The journey from here is yours. Pick a problem that matters to you, build
-            on the techniques you have learned, and contribute to the field. The most important chapters of
-            AI's story have not yet been written.
-        </p>
+        <p>In the next section, <a href="section-35.5.html" class="cross-ref">Section 35.5: Reliability Engineering for Agents Under Production Stress</a>, we turn to the engineering challenges of building agents that fail safely and recover gracefully under real-world conditions.</p>
     </div>
 
     <section class="bibliography">
@@ -500,7 +493,7 @@ <h2>Bibliography and Further Reading</h2>
 <nav class="chapter-nav">
     <a class="prev" href="section-35.3.html">Societal Impact and the Road Ahead</a>
     <a class="up" href="index.html">Chapter 35: AI, Society &amp; Open Problems</a>
-    <a class="next" href="../../appendices/appendix-a-mathematical-foundations/index.html">Mathematical Foundations for LLMs</a>
+    <a class="next" href="section-35.5.html">Reliability Engineering for Agents Under Production Stress</a>
 </nav>
 
     <footer>