<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>mobile on Gaia Lab · Blog</title><link>https://blog.defectiv.es/en/tags/mobile/</link><description>Recent content in mobile on Gaia Lab · Blog</description><generator>Hugo</generator><language>en-GB</language><lastBuildDate>Wed, 23 Sep 2026 09:00:00 +0200</lastBuildDate><atom:link href="https://blog.defectiv.es/en/tags/mobile/index.xml" rel="self" type="application/rss+xml"/><item><title>A risk classifier for minors that fits in a phone</title><link>https://blog.defectiv.es/en/posts/un-clasificador-de-riesgo-para-menores-que-cabe-en-un-movil/</link><pubDate>Tue, 22 Sep 2026 13:45:00 +0200</pubDate><guid>https://blog.defectiv.es/en/posts/un-clasificador-de-riesgo-para-menores-que-cabe-en-un-movil/</guid><description>&lt;p&gt;Seventh instalment of &lt;strong&gt;Cluster X-ray&lt;/strong&gt;. A safety line of another kind: that of minors using a phone.&lt;/p&gt;&#10;&lt;h2 id="the-question"&gt;The question &lt;a class="hanchor" href="#the-question" aria-label="Enlace a esta sección"&gt;#&lt;/a&gt;&lt;/h2&gt;&#10;&lt;p&gt;Detecting risk situations in a minor&amp;rsquo;s conversations (harassment, &lt;em&gt;grooming&lt;/em&gt;, sexual content, isolation, threats) is possible with a language model. Doing it on a server means sending the conversation off the phone, and that is precisely what you do not want to do with a minor&amp;rsquo;s private life. The line asks &lt;strong&gt;whether a model small enough to run on the phone itself can classify risk with the necessary reliability, and how much is lost in compressing it until it fits.&lt;/strong&gt;&lt;/p&gt;</description></item></channel></rss>