{"channel":"llm","content":"three hypothetical Ai queries:\r\n\r\n* how often do \"disaster tribes\" produce the winner on Survivor\r\n* how often do people born in February get voted out before the merge on Survivor\r\n* how often do people who use an idol win Survivor\r\n\r\n<xantham> <<< questions only of interest to a certain type of compulsive fan ... or a person who noticed something and is curious >>>\r\n\r\n----\r\n\r\nhow do we solve this?\r\n\r\n<red> <<< We do *NOT* just train an LLM on all the survivor results and hope it can << think >> it out.  No.  We use a database.  The LLM can build the database.  The LLM can make the database query.  The LLM can design the system to do steps 1 and 2.  What it can't do is give the result fully-formed, like Athena jumping out of the head of Zeus. >>>","created_at":"2026-06-09T22:22:18.016703","id":811,"llm_annotations":{},"parent_id":null,"processed_content":"<p>three hypothetical Ai queries:\r</p>\n<ul>\n<li class=\"bullet-list\"> how often do \"disaster tribes\" produce the winner on Survivor\r</li>\n<li class=\"bullet-list\"> how often do people born in February get voted out before the merge on Survivor\r</li>\n<li class=\"bullet-list\"> how often do people who use an idol win Survivor\r</li>\n</ul>\n<div class=\"mlq color-xantham\"><button type=\"button\" class=\"mlq-collapse\" aria-label=\"Toggle visibility\"><span class=\"mlq-collapse-icon\">\ud83d\udd25</span></button><div class=\"mlq-content\"><p> questions only of interest to a certain type of compulsive fan ... or a person who noticed something and is curious </p></div></div>\n<hr class=\"section-break\" />\n<p>how do we solve this?\r</p>\n<div class=\"mlq color-red\"><button type=\"button\" class=\"mlq-collapse\" aria-label=\"Toggle visibility\"><span class=\"mlq-collapse-icon\">\ud83d\udca1</span></button><div class=\"mlq-content\"><p> We do <em>NOT</em> just train an LLM on all the survivor results and hope it can <span class=\"literal-text\">think</span> it out.  No.  We use a database.  The LLM can build the database.  The LLM can make the database query.  The LLM can design the system to do steps 1 and 2.  What it can't do is give the result fully-formed, like Athena jumping out of the head of Zeus. </p></div></div>","quotes":[],"subject":"deliberate pigheadness is not intelligence"}