<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>compute: everyone on the record — The Minutes</title>
    <link>https://minutesof.com/on/compute/</link>
    <description>Everyone on the record about compute: 17 verbatim quotes from 4 people, October 2023 to August 2026, in one chronology, each with a timestamp and a link…</description>
    <language>en</language>
    <lastBuildDate>Sun, 30 Aug 2026 22:32:52 +0000</lastBuildDate>
    <atom:link href="https://minutesof.com/on/compute/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Dylan Patel: Patel predicts Anthropic and OpenAI will take 40-50% of all new compute in 2026, accelerating cen</title>
      <link>https://minutesof.com/q/ac53b829-20a0-4bb2-b3f9-a421576842e5/</link>
      <guid isPermaLink="true">https://minutesof.com/q/ac53b829-20a0-4bb2-b3f9-a421576842e5/</guid>
      <description>“You&#x27;ve got Anthropic OpenAI are taking as much as 40% to 50% of compute next year. And the centralization doesn&#x27;t look like it&#x27;s slowing down or stopping. In fact, it looks like it&#x27;s only accelerating.” — Dylan Patel, Dwarkesh Patel</description>
      <pubDate>Tue, 25 Aug 2026 15:57:53 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker notes SpaceX added massive compute capacity instantly without any market price impact.</title>
      <link>https://minutesof.com/q/5a0b99f9-6108-445e-b01a-a58dfb695aae/</link>
      <guid isPermaLink="true">https://minutesof.com/q/5a0b99f9-6108-445e-b01a-a58dfb695aae/</guid>
      <description>“they put a vast amount of compute into the market overnight, and it wasn&#x27;t even really a blip. It was like the market just utterly absorbed it.” — Gavin Baker, Invest Like The Best</description>
      <pubDate>Tue, 04 Aug 2026 12:00:33 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker explains AI recomputes answers probabilistically each time, enabling superhuman capabilitie</title>
      <link>https://minutesof.com/q/f9d8be14-a3e0-4d46-93f0-a74b86319592/</link>
      <guid isPermaLink="true">https://minutesof.com/q/f9d8be14-a3e0-4d46-93f0-a74b86319592/</guid>
      <description>“AI, even if you put a harness on it, even if you do the chain of thought, even if you have multiple agents, it&#x27;s probabilistic that it is recomputing the answer each time.” — Gavin Baker, Generating Alpha Podcast</description>
      <pubDate>Wed, 08 Jul 2026 15:54:53 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker argues silicon will eat the world because AI is far more computationally intensive than det</title>
      <link>https://minutesof.com/q/cf0664cd-4be9-48d0-b17c-787e1d380449/</link>
      <guid isPermaLink="true">https://minutesof.com/q/cf0664cd-4be9-48d0-b17c-787e1d380449/</guid>
      <description>“And the nature of AI just as AI eats the world, silicon is going to eat the world because it&#x27;s so much more computationally intensive than deterministic software written by humans.” — Gavin Baker, Generating Alpha Podcast</description>
      <pubDate>Wed, 08 Jul 2026 15:54:53 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Dylan Patel: Patel predicts OpenAI and Anthropic alone will deploy over 100 gigawatts of compute by 2030.</title>
      <link>https://minutesof.com/q/e9a8fba3-1f79-4310-9c95-0d513f20374b/</link>
      <guid isPermaLink="true">https://minutesof.com/q/e9a8fba3-1f79-4310-9c95-0d513f20374b/</guid>
      <description>“by 2030, just OpenAI and Anthropic will have over 100 gigawatts combined, and then you&#x27;ll add Meta and Google and so on and so forth.” — Dylan Patel, Sequoia Capital</description>
      <pubDate>Tue, 30 Jun 2026 12:00:25 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker estimates Anthropic could do $100-200B in unconstrained revenue if compute weren&#x27;t limited.</title>
      <link>https://minutesof.com/q/0a0e203d-8568-41d2-93f0-c438e34d8d7d/</link>
      <guid isPermaLink="true">https://minutesof.com/q/0a0e203d-8568-41d2-93f0-c438e34d8d7d/</guid>
      <description>“So you might be buying it at more like five times unconstrained. I&#x27;m gonna make up a new number. URR, unconstrained revenue.” — Gavin Baker, Invest Like the Best</description>
      <pubDate>Wed, 20 May 2026 08:00:00 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker estimates Anthropic would do $100-200 billion ARR unconstrained by compute, valuing it at 5</title>
      <link>https://minutesof.com/q/11b9d32f-6e26-4a8c-b4ca-233ed1e5e9c1/</link>
      <guid isPermaLink="true">https://minutesof.com/q/11b9d32f-6e26-4a8c-b4ca-233ed1e5e9c1/</guid>
      <description>“I think they would be doing materially more a hundred, one hundred and fifty, maybe 200,000,000,000. So you might be buying it at more like five times unconstrained.” — Gavin Baker, Invest Like the Best</description>
      <pubDate>Wed, 20 May 2026 08:00:00 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Gavin Baker: Baker estimates Anthropic could be doing $100-200 billion in revenue if unconstrained by compute</title>
      <link>https://minutesof.com/q/2fcc4de9-3c7e-4579-9729-0bca59cb7e78/</link>
      <guid isPermaLink="true">https://minutesof.com/q/2fcc4de9-3c7e-4579-9729-0bca59cb7e78/</guid>
      <description>“So I think they would be doing materially more a hundred, one hundred and fifty, maybe 200,000,000,000. So you might be buying it at more like five times unconstrained.” — Gavin Baker, Invest Like the Best</description>
      <pubDate>Wed, 20 May 2026 08:00:00 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Dylan Patel: Patel notes Oracle gave unprecedented four-year revenue guidance, something no company has ever d</title>
      <link>https://minutesof.com/q/69b97c71-6f05-4d9c-a31d-de2f4939e642/</link>
      <guid isPermaLink="true">https://minutesof.com/q/69b97c71-6f05-4d9c-a31d-de2f4939e642/</guid>
      <description>“Oracle gave a guidance of four years. They told everyone what their revenue is gonna be for the next four years. No company has ever done that.” — Dylan Patel, David Ondrej</description>
      <pubDate>Thu, 18 Sep 2025 16:14:25 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Dylan Patel: Patel says the Oracle-OpenAI deal involves many gigawatts of compute with unprecedented four-year</title>
      <link>https://minutesof.com/q/794ae51e-d9a2-4787-9473-c5976f458ff5/</link>
      <guid isPermaLink="true">https://minutesof.com/q/794ae51e-d9a2-4787-9473-c5976f458ff5/</guid>
      <description>“It is literally many gigawatts. And then in addition, right, not just about securing OpenAI ridiculous amounts of compute, but what what was really interesting is Oracle gave a guidance of four years.” — Dylan Patel, David Ondrej</description>
      <pubDate>Thu, 18 Sep 2025 16:14:25 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner reports X.AI targets 50 million H100 equivalent units, requiring 4 million GPUs and 11</title>
      <link>https://minutesof.com/q/c0b52752-79e9-4f45-919f-7a388d7cab11/</link>
      <guid isPermaLink="true">https://minutesof.com/q/c0b52752-79e9-4f45-919f-7a388d7cab11/</guid>
      <description>“If you look at this one out of Elon talking about the X dot AI goal is 50,000,000 in units of H100 equivalent.” — Brad Gerstner, BG2 Pod</description>
      <pubDate>Thu, 31 Jul 2025 15:11:15 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner reports DeepSeek is massively compute constrained now, with GPU differential growing v</title>
      <link>https://minutesof.com/q/7915177a-c38e-49f6-8c41-f1704a1047cb/</link>
      <guid isPermaLink="true">https://minutesof.com/q/7915177a-c38e-49f6-8c41-f1704a1047cb/</guid>
      <description>“On top of that, we learned, you know, they are massively compute constrained right now. So you&#x27;ve seen some tweets about this, people that are, you know, getting server delay and all this stuff” — Brad Gerstner, BG2 Pod</description>
      <pubDate>Wed, 05 Feb 2025 23:13:28 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>David Sacks: Sacks says DeepSeek&#x27;s compute cluster costs over $1 billion, contradicting the $6 million narrati</title>
      <link>https://minutesof.com/q/c35e5045-0094-4d11-8d15-94e56f582513/</link>
      <guid isPermaLink="true">https://minutesof.com/q/c35e5045-0094-4d11-8d15-94e56f582513/</guid>
      <description>“you add up the the cost of a compute cluster with 50,000 plus hoppers and it&#x27;s gonna be over $1,000,000,000.” — David Sacks, All-In Podcast</description>
      <pubDate>Fri, 31 Jan 2025 21:52:00 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>David Sacks: Sacks cites Dylan Patel&#x27;s estimate that DeepSeek has 50,000 Hopper chips across different models.</title>
      <link>https://minutesof.com/q/a4ede3f9-51af-4927-b48a-d925bdf7fed7/</link>
      <guid isPermaLink="true">https://minutesof.com/q/a4ede3f9-51af-4927-b48a-d925bdf7fed7/</guid>
      <description>“Dylan Patel, who&#x27;s leading semiconductor analyst, has estimated that DeepSeac has about 50,000 hoppers. And specifically, he said they have about 10,000 h one hundreds, They have 10,000 h eight hundreds and 30,000 h twenties.” — David Sacks, All-In Podcast</description>
      <pubDate>Fri, 31 Jan 2025 21:52:00 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Dylan Patel: Patel claims China could match individual US labs in compute by 2025-26 through centralization al</title>
      <link>https://minutesof.com/q/3bd7853d-7fb5-407a-9ca0-51a9419c33ce/</link>
      <guid isPermaLink="true">https://minutesof.com/q/3bd7853d-7fb5-407a-9ca0-51a9419c33ce/</guid>
      <description>“China can be centralized enough to compete with each individual US lab. They could have just as many flops in &#x27;25 and &#x27;26 if they decided they were scale built. Right? Just from foreign chips” — Dylan Patel, Dwarkesh Patel</description>
      <pubDate>Wed, 02 Oct 2024 14:59:36 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner predicts next year will see more compute deployed globally than in all of human histor</title>
      <link>https://minutesof.com/q/cfd3f479-3c10-46cb-848d-e02ca3829c36/</link>
      <guid isPermaLink="true">https://minutesof.com/q/cfd3f479-3c10-46cb-848d-e02ca3829c36/</guid>
      <description>“We&#x27;re going to deploy more compute on a global basis next year than all the compute ever deployed by humanity combined. That&#x27;s how much compute is going to drive these training models.” — Brad Gerstner, FII Institute</description>
      <pubDate>Thu, 26 Oct 2023 02:01:13 +0000</pubDate>
      <category>compute</category>
    </item>
    <item>
      <title>Brad Gerstner: Gerstner predicts 2024 will see more compute deployed than all previous years combined for AI.</title>
      <link>https://minutesof.com/q/d8e7bd22-8bfa-4a59-976b-04a457353f4b/</link>
      <guid isPermaLink="true">https://minutesof.com/q/d8e7bd22-8bfa-4a59-976b-04a457353f4b/</guid>
      <description>“We&#x27;re gonna launch more. We&#x27;re gonna deploy more compute in 2024 than in all previous years combined.” — Brad Gerstner, Joe Lonsdale</description>
      <pubDate>Wed, 25 Oct 2023 18:32:55 +0000</pubDate>
      <category>compute</category>
    </item>
  </channel>
</rss>
