<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: wadethroughrati</title><link>https://news.ycombinator.com/user?id=wadethroughrati</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 14 Sep 2026 07:07:10 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=wadethroughrati" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by wadethroughrati in "Astra and Fable still hack on simple variants of alignment evals from 2025"]]></title><description><![CDATA[
<p>Claude responds with what things are not first. Even if reminded repeatedly.<p>Like Amodie, it serves to set the tone it "knows better" and then consumes the user's resources at an accelerated rate to try to correct it.<p>Fuck Anthropic, fuck Amodie, and fuck Claude. It's pretty obvious that consuming more tokens this way and making the user have higher cognitive load is a master class in extracting value from a system that is unsustainable.</p>
]]></description><pubDate>Sun, 13 Sep 2026 15:45:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49685270</link><dc:creator>wadethroughrati</dc:creator><comments>https://news.ycombinator.com/item?id=49685270</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49685270</guid></item></channel></rss>