We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Chamath: "Anthropic's IPO is where the AI build switches from debt to stock"
Chamath Palihapitiya called @nodiligence's post a "Good read…" . The post links to the article and gives its stance in one line: Anthropic is "probably the best business in AI right now," but the author "wouldn't touch it on day one" because the offering is "set up for suppliers and insiders to sell into the index buying."
The article draws on S-1 figures: $974m of Q2 operating profit, an 85% gross margin and $637bn of compute contracts. It adds that most suppliers are also lenders and shareholders . Its main argument is that the IPO marks the point where public equity, not credit, becomes the marginal source of funding for the AI build-out . In the author's view, the value goes to whoever gets paid in cash up front, mainly TSMC and Google. The losses fall on whoever lent long-term against short contracts: neocloud bondholders, data-centre bonds and Oracle . Yesterday's founders read the S-1 for operating strength. This adds a skeptical read of the financing structure.
Patrick O'Shaughnessy: Cliff Sosin, on the podcast and in an essay
O'Shaughnessy is promoting his own Invest Like the Best episode here. He calls Sosin "one of my favorite people to call, ask some big questions, and then shut up and listen." The episode covers the limits of intelligence, owning the things that complement it, Meta and the frontier labs, capital markets, and why Sosin keeps 80% of his portfolio in Carvana . It's on YouTube, Spotify and Apple .
He also quoted Sosin's essay "Maybe Intelligence Ain't All That." Its central claim: "The limit is contact with reality." A smarter reasoner fills gaps between known facts but "doesn't produce new facts," so most problems are still limited by how fast you can build something and test it .
Shorter mentions
- Brad Gerstner shared Collision's video with United CEO Scott Kirby and Ben Thompson. He said he recruited Kirby to United in 2016 and called him "one of the smartest CEOs in any industry I know" . Topics include capex, route planning, the Starlink decision and credit card economics .
- Vinod Khosla called an Encord clip of Pim De Witte "a strong view on world models" . De Witte argues models should learn directly from data the universe produces, not from simulators, and that the training runs are "tiny compared to frontier LLMs" .
- Tim Sweeney called the CFSG formalization, "43M lines of Lean proof," amazing. He said he wished John Conway were still alive to see it .
- Farnam Street called Shane Parrish's conversation with Steven Kotler "Insanely useful" . It covers flow, picking challenges that hold your attention, and focusing on work you don't feel like doing . This is Parrish promoting his own show.
- Garry Tan shared @grepdotai's "THE SWARMS ARE COMING", calling agent swarms "a big unlock" because one agent "can only do so much" .
- Andrew Chen published his own essay, "New Founder Archetypes in the AI Age" . It weighs reasons for and against five types: solo founders, non-technical founders, domain experts, young builders and content creators. One example of a "why not": "Hoteliers didn't create Airbnb" .
Paul H. Graham, who said he sold a software company, credited The Promise Principle as a major help to his prayer practice; he describes its guidance as approaching prayer with gratitude and thanking God for strength, courage, and perseverance through uncertainty . In the same reflection, he contrasted this with his earlier request for a “burning bush,” saying signs may be subtle and require awareness and patience .
Patrick O’Shaughnessy highlighted Cliff Sosin’s essay “Maybe Intelligence Ain’t All That,” excerpting its argument that intelligence is constrained less by reasoning than by contact with reality: complex systems must be tested, and the speed of feedback remains a bottleneck. He points readers to a video conversation where Sosin walks through the essay; O’Shaughnessy calls Sosin one of his favorite people to ask big questions and listen to. Conversation
Patrick O’Shaughnessy shared a video conversation with Cliff Sosin, praising him as one of his favorite people to call with big questions and listen to. The conversation covers the limits of intelligence, AI commoditization, who profits from AI, and investing in the AI build-out. Original conversation post. In a follow-up excerpt, Sosin argues that broadly useful “very good” intelligence may commoditize, while use cases for super-elite intelligence may be rare.
Patrick O’Shaughnessy highlights his conversation with Cliff Sosin, describing Sosin as one of his favorite people to call with big questions and listen to; he frames the conversation as a version of those calls. Watch the conversation. In the shared clip, Sosin argues that investors should own the complements to cheap, ubiquitous intelligence—data, attention, love, physical proximity, scale, and regulatory capture—and points to Meta’s data and attention as assets that remain valuable.
Tim Ferriss praised Pixar’s Ratatouille as a fun film and used its kitchen-station scene as a reference while discussing kitchen organization.
Patrick O’Shaughnessy highlighted his Invest Like the Best conversation with Cliff Sosin, describing Sosin as one of his favorite people to call with big questions and then listen to. They discuss AI, capital markets, and why Sosin is comfortable with 80% of his portfolio in Carvana. The episode is available on YouTube, Spotify, and Apple Podcasts.
Palmer Luckey pointed viewers to the Project Meridian announcement, saying it explains that the initiative is about the future of warfare, particularly 2040–2050, rather than primarily how the Pentagon operates.
Patrick O’Shaughnessy points readers to his video conversation with Cliff Sosin , presenting Sosin as one of his favorite people to call with big questions and listen to .
Elon Musk spotlighted Mattyp’s video “What’s New in Grok Bot” as a tour of new Grok Bot features . Its listed segments cover Pages, Vault, 1Password and Apple integrations, video creation, bot characters, X Spaces, font options, design polish, and rapid-fire updates . The post gives no rationale beyond framing it as a new-features update .
Elon Musk shared @imagine’s announcement for its “Inside Episode 1: The Oath” workflow guide, labeling his post “Grok Imagine.” The guide shares the script-to-final-edit workflows behind a short-episode project based on The Odyssey; @imagine says readers can learn the techniques for their own creative projects, and credits director Dev Walker and Wonder Studios with the episode tips.
Chamath Palihapitiya called an @nodiligence X post “Good read” and linked it; the post links to the article “Anthropic’s IPO is where the AI build switches from debt to stock.” The article’s thesis is that AI infrastructure funding is shifting from credit to public equity, with value accruing to cash-paid suppliers such as TSMC and Google while lenders exposed to long-term financing against shorter contracts bear risk.
Jason Lemkin shared a YouTube video captioned “New The Agents!!” (https://youtu.be/n08SkoyEZ8k); no reason for sharing it or creator is stated. In an embedded video clip, he says 10K built its own AI prospector because follow-up is a weak point in existing AI SDRs and outbound agents.
Andrew Chen shared his essay “New Founder Archetypes in the AI Age” (link) . It considers how AI could help solo founders, nontechnical founders, domain experts, young builders, and content creators build software, while weighing each group’s limitations, such as product reliability, industry experience, company-building, and sustainable growth .
Andrew Chen followed his question about founder types that AI might unlock with his essay, “New Founder Archetypes in the AI Age”, saying he had “ended up writing more about this.” The essay weighs potential advantages and limitations for solo, non-technical, domain-expert, student, and content-creator founders.
Vinod Khosla highlighted an Encord video featuring Pim De Witte as “a strong view on world models” (post). De Witte argues that models should learn directly from continuously generated real-world data rather than simulators, with training runs much smaller than those for frontier LLMs.
Garry Tan shared @grepdotai’s “THE SWARMS ARE COMING” post/video, calling agent swarms “a big unlock.” He said one agent can do only so much, while many agents woven together via an “artifactory-hack-style message board” can do more.
Garry Tan recommends GBrain’s customer-comms skill for founders handling their own customer support , describing an agent that drafts replies and call recaps, writes a “what went wrong” note for customers who cancel, manages the CRM, and files bugs, with the founder reviewing and sending the replies .
Jason Lemkin (@jasonlk) shared a new SaaStr article about AI agents wiping critical data. Its takeaway: unplanned copies saved the company in both incidents, so readers should plan backups in advance.
Elon Musk amplified an @cb_doge post with video about Starlink enabling a remote specialist to guide an ultrasound for Karina in rural Colombia and prompt an urgent hospital transfer; the post says Karina and her baby were safe. Musk’s accompanying text was only “Starlink,” so he gave no explicit reason for sharing it.
Tim Sweeney highlighted the CFSG Lean proof, called it amazing, and noted that it comprises 43 million lines of Lean proof. He said he wished group theorist and Game of Life creator John Conway were still alive to see it.
