<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Bugs on The Coders Blog</title><link>https://thecodersblog.com/tag/bugs/</link><description>Recent content in Bugs on The Coders Blog</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 29 Apr 2026 21:11:43 +0000</lastBuildDate><atom:link href="https://thecodersblog.com/tag/bugs/index.xml" rel="self" type="application/rss+xml"/><item><title>Anthropic's $200 Bug: When AI API Errors Cost You, and Refunds Are Denied</title><link>https://thecodersblog.com/hermes-md-anthropic-s-billing-bug-refused-refused-refunds-and-the-cost-of-trust-2026/</link><pubDate>Wed, 29 Apr 2026 21:11:43 +0000</pubDate><guid>https://thecodersblog.com/hermes-md-anthropic-s-billing-bug-refused-refused-refunds-and-the-cost-of-trust-2026/</guid><description>&lt;p&gt;You thought your AI API usage was covered by your subscription. Then, a silent bug routed it to &amp;rsquo;extra usage&amp;rsquo;, costing hundreds, with refunds denied. Let&amp;rsquo;s talk about why Anthropic&amp;rsquo;s &lt;strong&gt;&amp;lsquo;HERMES.md&amp;rsquo; blunder&lt;/strong&gt; isn&amp;rsquo;t just a technical glitch, but a stark warning about the future of AI billing and provider accountability.&lt;/p&gt;
&lt;h2 id="the-financial-black-box-when-ai-costs-become-a-gamble"&gt;The Financial Black Box: When AI Costs Become a Gamble&lt;/h2&gt;
&lt;p&gt;The allure of AI APIs, with their promise of unparalleled capabilities, often casts a long shadow over the prosaic yet critical reality of their pricing models. Developers and FinOps teams are implicitly paying a &lt;strong&gt;&amp;ldquo;cost of trust&amp;rdquo;&lt;/strong&gt;—a blind faith that the vendor&amp;rsquo;s billing mechanisms are transparent and accurate. This faith, as we&amp;rsquo;ve seen, is often misplaced.&lt;/p&gt;</description></item></channel></rss>