<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Synthetic-Data on Local First AI</title><link>https://localfirstai.eu/tags/synthetic-data/</link><description>Recent content in Synthetic-Data on Local First AI</description><image><title>Local First AI</title><url>https://localfirstai.eu/images/og-default.png</url><link>https://localfirstai.eu/images/og-default.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 20 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://localfirstai.eu/tags/synthetic-data/index.xml" rel="self" type="application/rss+xml"/><item><title>Two AIs Disagreed on 1 in 5 Test Enquiries. We Deleted Those.</title><link>https://localfirstai.eu/posts/2026-10-20-synthetic-data-lessons/</link><pubDate>Tue, 20 Oct 2026 00:00:00 +0000</pubDate><guid>https://localfirstai.eu/posts/2026-10-20-synthetic-data-lessons/</guid><description>One AI wrote our test enquiries, a second AI labelled them again, and we deleted the ones they disagreed on: nearly one in five. Then we scored our model on the rest. That is one of seven things synthetic data did to our experiments since September, each with the record it comes from. Most point the same way: synthetic data is cleaner, easier and more like its writer than the world it stands in for.</description></item></channel></rss>