<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>ai models: everyone on the record — The Minutes</title>
    <link>https://minutesof.com/on/ai-models/</link>
    <description>Everyone on the record about ai models: 7 verbatim quotes from 4 people, February 2024 to May 2026, in one chronology, each with a timestamp and a link to…</description>
    <language>en</language>
    <lastBuildDate>Sun, 30 Aug 2026 22:32:47 +0000</lastBuildDate>
    <atom:link href="https://minutesof.com/on/ai-models/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Gavin Baker: Baker says understanding frontier AI now requires enterprise usage-based plans, not consumer subs</title>
      <link>https://minutesof.com/q/7b49081d-b3f9-445d-aba4-d290d5374ebd/</link>
      <guid isPermaLink="true">https://minutesof.com/q/7b49081d-b3f9-445d-aba4-d290d5374ebd/</guid>
      <description>“To understand what Frontier AI is capable of today, even for a non coding use case, need to have Cloud Code or Codex five point Codex. And you need to be on an enterprise plan.” — Gavin Baker, Invest Like the Best</description>
      <pubDate>Wed, 20 May 2026 08:00:00 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner notes Alibaba&#x27;s Qwen model has passed 400 million downloads under Apache 2.0 license.</title>
      <link>https://minutesof.com/q/61b701ab-5440-4bd6-b877-eb17aff4e766/</link>
      <guid isPermaLink="true">https://minutesof.com/q/61b701ab-5440-4bd6-b877-eb17aff4e766/</guid>
      <description>“Of course, that&#x27;s released like these other models under the Apache 2.0 open source license, so very open, as Bill&#x27;s talked about.” — Brad Gerstner, BG2 Pod</description>
      <pubDate>Thu, 31 Jul 2025 15:11:15 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Dylan Patel: Patel reports AI models improved from below 4% to 70% accuracy on Frontier Math in six months acr</title>
      <link>https://minutesof.com/q/2bb1c7b0-0d8f-45e8-a0a1-db8ef64742e6/</link>
      <guid isPermaLink="true">https://minutesof.com/q/2bb1c7b0-0d8f-45e8-a0a1-db8ef64742e6/</guid>
      <description>“frontier math. And the models went from sub 4% to 70% in six months. Right? And this is from not just OpenAI, but also Anthropic and Google and DeepSeek.” — Dylan Patel, Going Direct</description>
      <pubDate>Sun, 22 Jun 2025 21:30:00 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Dylan Patel: Patel reports AI models improved from under 4% to 70% on Frontier Math in six months.</title>
      <link>https://minutesof.com/q/9274e9a4-6baa-4e2a-b44a-3c2207f42c82/</link>
      <guid isPermaLink="true">https://minutesof.com/q/9274e9a4-6baa-4e2a-b44a-3c2207f42c82/</guid>
      <description>“Like, can&#x27;t do all the math, all the problems in frontier math. And the models went from sub 4% to 70% in six months.” — Dylan Patel, Going Direct</description>
      <pubDate>Sun, 22 Jun 2025 21:30:00 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Dylan Patel: if they put them all in one cluster, they could have a bigger model than any of the labs next yea</title>
      <link>https://minutesof.com/q/46a70da8-5378-46a9-a527-ee41c13e3fa8/</link>
      <guid isPermaLink="true">https://minutesof.com/q/46a70da8-5378-46a9-a527-ee41c13e3fa8/</guid>
      <description>“if they put them all in one cluster, they could have a bigger model than any of the labs next year. Right?” — Dylan Patel, Dwarkesh Patel</description>
      <pubDate>Wed, 02 Oct 2024 14:59:36 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Bill Gurley: Gurley says Tesla replaced coded self-driving with neural network using driver videos as input an</title>
      <link>https://minutesof.com/q/0095dec9-4639-4988-8110-944d679ce57e/</link>
      <guid isPermaLink="true">https://minutesof.com/q/0095dec9-4639-4988-8110-944d679ce57e/</guid>
      <description>“What we learned this week is that they&#x27;ve completely tossed all of that out and gone with a neural network model where they&#x27;re uploading videos from their best drivers.” — Bill Gurley, BG2 Pod</description>
      <pubDate>Thu, 07 Mar 2024 16:29:00 +0000</pubDate>
      <category>ai models</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner predicts Anthropic models will match GPT-4 and 4.5 turbo quality in the near term.</title>
      <link>https://minutesof.com/q/69e99490-3061-434b-a68c-846f3bb5b2fb/</link>
      <guid isPermaLink="true">https://minutesof.com/q/69e99490-3061-434b-a68c-846f3bb5b2fb/</guid>
      <description>“They&#x27;re gonna have models which are every bit as good, you know, in the near term as chat GPT four and four point five turbo, etcetera.” — Brad Gerstner, CNBC Television</description>
      <pubDate>Fri, 02 Feb 2024 18:54:25 +0000</pubDate>
      <category>ai models</category>
    </item>
  </channel>
</rss>
