<?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>hi, it&#39;s mike</title>
    <link>https://mike.puddingtime.org/tags/sabotage/</link>
    <description>Recent content on hi, it&#39;s mike</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en-us</language>
    <managingEditor>mike@puddingtime.org (mike)</managingEditor>
    <webMaster>mike@puddingtime.org (mike)</webMaster>
    <copyright>© 2026, mike</copyright>
    <lastBuildDate>Tue, 30 Jan 2024 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://mike.puddingtime.org/tags/sabotage/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Daily Notes for 2024-01-30</title>
      <link>https://mike.puddingtime.org/posts/2024-01-30-daily-notes/</link>
      <pubDate>Tue, 30 Jan 2024 00:00:00 +0000</pubDate><author>mike@puddingtime.org (mike)</author>
      <guid>https://mike.puddingtime.org/posts/2024-01-30-daily-notes/</guid>
      <description>The AI I am supposed to babysit insists that I just ate a sundae.</description>
      <content:encoded><![CDATA[<h2 id="sure-i-d-be-happy-to-help-train-your-model">Sure, I&rsquo;d be happy to help train your model</h2>
<p>My continuous glucose monitor works with an app. The app is mostly cool: Good historical data, great for developing day-to-day awareness, and it&rsquo;s hooked into other stuff so I can make connections between sleep, exercise, and blood sugar.</p>
<p>It also has a meal diary. When I first started using the app, you could choose between a photo entry or a detailed entry for each meal. If you chose a photo entry, it&rsquo;d just let you take a picture of your meal and timestamp it. If you felt like going back and filling in what you ate, it would provide a link to a database on kinds of food and their potential metabolic impacts. If you chose a detailed entry, the picture became optional and you could fill in more detail on the spot.</p>
<p>I preferred photo entries. I know what&rsquo;s in a given meal so it was fine to look at a week of historical data and check in on photos of the meals that seemed to be connected to spikes or dips.</p>
<p>Recently the app succumbed to the AI fad, so the photo log UI changed: Now the app tries to guess what you&rsquo;re eating by analyzing the photo. I don&rsquo;t think, in over a month of meals, it has had a better than 10 percent success rate. Some of its guesses are simply wild. It isn&rsquo;t learning from me specifically, because I have a very consistent breakfast choice and no amount of correcting the same basic picture taken almost daily has swayed the app.</p>
<p>There&rsquo;s no way to turn the AI assistant off, though. They wedged it into the workflow and your choices are &ldquo;correct it&rdquo; or &ldquo;ignore it.&rdquo;</p>
<p>I tried to correct it for about a week then snapped out of it: There is no way the damn thing made it out of beta without everyone realizing it is infuriatingly bad at what it does. I&rsquo;m sure they intentionally released this half-baked thing counting on people to train it by correcting it for the sake of accurate records.</p>
<p>So I just accept whatever it proposes. You think that bowl of lentil soup is a chocolate sundae? Sure. Yes, now that you mention, I thought I was eating a tasty omelette, but I can see now that it was a ham.</p>
<p>I&rsquo;m happy to send them an invoice if they&rsquo;d prefer different behavior on my part.</p>
]]></content:encoded>
    </item>
  </channel>
</rss>
