https://blog.without.hosting/tags/hot-takes/Recent content in Hot Takes on Without hostingHugo -- gohugo.ioen-USThu, 18 Jun 2026 10:36:22 +0000https://blog.without.hosting/posts/backspine-dont-read-agent-text/Thu, 18 Jun 2026 10:36:22 +0000https://blog.without.hosting/posts/backspine-dont-read-agent-text/<p>Stop reading what your coding agent says. Read what it produces.</p> <p>Coding agents are here to produce code. Not chatter about it. The textual output — the stream of &ldquo;I&rsquo;ll fix this by doing X, then Y, then Z&rdquo; — is noise.</p> <h2 id="the-bad-feedback-loop">The bad feedback loop</h2> <p>Instead of evaluating the agent&rsquo;s output (the code, the diff, the structural changes), you evaluate what it tells you about its output. The agent sounds confident. You approve. The code is wrong.</p>https://blog.without.hosting/posts/x-ai-quote-fabrication/Sun, 07 Jun 2026 00:50:00 +0000https://blog.without.hosting/posts/x-ai-quote-fabrication/<blockquote> <p><strong>Status: live, ongoing.</strong> The account is a systematic content farm posting 2–3 fabricated quotes daily. The DAO behind it is a ~5-month-old shell with unverifiable leadership and a false &ldquo;Official Polymarket Community&rdquo; claim. Last updated: 2026-06-08.</p> </blockquote> <p>I keep seeing videos on X attributed to &ldquo;An Anthropic engineer&rdquo; or &ldquo;An OpenAI researcher&rdquo; giving a 20–30 minute talk about agent reliability, prompt injection, or some other sharp operational topic. The quotes that come out of them are perfect. They sound like the inside view. They are also extremely shareable.</p>