<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Machine-Learning on AegisGate — Secure Every AI Interaction</title><link>https://aegisgatesecurity.io/tags/machine-learning/</link><description>Recent content in Machine-Learning on AegisGate — Secure Every AI Interaction</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Sun, 20 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://aegisgatesecurity.io/tags/machine-learning/feed.xml" rel="self" type="application/rss+xml"/><item><title>From Shadow to Shield: Validating 8.5M Requests and Flipping Our Detectors to Blocking</title><link>https://aegisgatesecurity.io/blog/from-shadow-to-shield-8-5m-requests-blocking-mode/</link><pubDate>Sun, 20 Sep 2026 00:00:00 +0000</pubDate><guid>https://aegisgatesecurity.io/blog/from-shadow-to-shield-8-5m-requests-blocking-mode/</guid><description>In our previous post, we analyzed real-world AI attack patterns from the last 90 days, ran them against our own detection stack, and found a 52.32% detection rate. We fixed six blind spots in our L1 regex patterns, achieved 100% on the adversarial test suite, and shipped v4.5.0.
That was Act I — the regex gaps. Act II was bigger.
The Question We Couldn&amp;rsquo;t Answer v4.5.0 shipped with five advanced detectors all in alert-only mode.</description></item></channel></rss>