http://programmer.ie/tags/llm-judge/ Recent content in LLM Judge on Programmer.ie: Modern AI programming Hugo en-us Sat, 29 Aug 2026 23:14:00 +0100 http://programmer.ie/books/hallucination-from-first-principles/04-chapter/ Sat, 29 Aug 2026 23:14:00 +0100 http://programmer.ie/books/hallucination-from-first-principles/04-chapter/ <p>The first three chapters deliberately avoided building a detector.</p> <p>Before designing one, we needed to define exactly what it would be expected to detect.</p> <p>Chapter 1 established that fluent generation can continue after evidential support has weakened or disappeared.</p> <p>Chapter 2 showed that <em>hallucination</em> covers several different failure relationships.</p> <p>Chapter 3 then separated the objects a reliability system must not collapse:</p> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-text" data-lang="text"><span style="display:flex;"><span>truth </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>evidence </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>support </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>attribution </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>provenance </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>verification </span></span><span style="display:flex;"><span>≠ </span></span><span style="display:flex;"><span>policy acceptance </span></span></code></pre></div><p>Now we can finally ask the engineering question:</p>