<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>What I'm Learning from My AI Experiments on How Marketing Technology Works®</title><link>https://howmarketingtechnology.works/series/ai-experiments/</link><description>Recent content in What I'm Learning from My AI Experiments on How Marketing Technology Works®</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Sun, 11 Oct 2026 12:00:00 +0000</lastBuildDate><atom:link href="https://howmarketingtechnology.works/series/ai-experiments/index.xml" rel="self" type="application/rss+xml"/><item><title>My AI Guardrails Held Only When a Script Checked the Agent's Work</title><link>https://howmarketingtechnology.works/ai-guardrails-instructions-vs-scripts/</link><pubDate>Sun, 11 Oct 2026 12:00:00 +0000</pubDate><guid>https://howmarketingtechnology.works/ai-guardrails-instructions-vs-scripts/</guid><description>&lt;h2 id="what-this-series-is"&gt;What this series is&lt;/h2&gt;&#10;&lt;p&gt;I run my consulting business with help from AI agents I built myself in Claude Code. They research companies, keep track of my projects, and work with me on my writing. I set the angle, the sources, and the rules before anything gets drafted, we write together, and I do the final read.&lt;/p&gt;&#10;&lt;p&gt;&lt;em&gt;What I&amp;rsquo;m Learning from My AI Experiments&lt;/em&gt; is where I share what broke. Each post covers one experiment: the problem, what I wanted, what I did, what happened, how I fixed it, whether that worked, and what a marketing team should take from it. The numbers come from my own test logs. They&amp;rsquo;re small samples from one person&amp;rsquo;s setup, so treat them as something to test in your own environment.&lt;/p&gt;</description></item></channel></rss>