Free — sign in with your email, no password.
Podcasts on 26 October 2026
A free-wandering conversation arguing that outcomes hinge on a few dominant factors, that simple ideas taken seriously beat a less-committed field, and that taste, product quality and attention shape what wins.
12 key points · 2 quotesMatt Murphy explains how Menlo built conviction, structure and flexibility to back Anthropic early, and how AI's speed, durable enterprise demand and team-based investing are reshaping venture.
12 key pointsKevin Roose argues that AI labs are driven by a small, zealous, belief-fueled culture betting on LLMs scaling to AGI, while Eli Patel remains skeptical because of the models' brittleness.
12 key points · 1 quoteNathaniel walks through his AI stack by use case and finds that personal agents like Dot and Duet are, so far, valuable mainly for proactive work rather than as a replacement for his main tools.
10 key pointsComedian Taylor Tomlinson explains how stand-up works as a writing practice: mine strong feelings, test material live, know how audiences see you, and keep writing through self-doubt.
12 key points · 2 quotesInstagram head Adam Mosseri explains the risky push into long-form and horizontal video, why creator payments must break even, how ranking works, and why AI anger is growing.
11 key pointsSimile's founder June argues that human behavior can be simulated from grounded data on real people, predicting it with 85%+ accuracy, and that these simulations let companies test decisions before making them.
10 key pointsSatya Nadella argues that because alignment doesn't fall out of scaling, firms should build zero-trust containment around models and own their intelligence through private evals, while AI's real payoff is inflecting world GDP growth.
9 key pointsMarvin Schwaibold argues AI makes designers editors and full-stack builders who prototype in code, while good web design still depends on one clear focus, reduction, progressive disclosure, and respect for native interaction.
11 key points · 3 quotesOpenAI's Ari and James argue computer use is approaching superhuman speed, works universally without integrations, and now needs enterprise controls, better UX and proactivity to reach mainstream consumers and businesses.
8 key pointsNick Kokonas says every arbitrage is really a time-scale gap, and that his moves from trading to restaurants to software to magic came from seeking discomfort, doing the daily grind, and borrowing from adjacent fields.
12 key pointsReddit built a custom prototyping harness on its design system, real data and an MCP, so designers and PMs can make production-like prototypes quickly and hand them off to engineering.
9 key pointsMark Cuban argues that success comes from mastering one skill and out-preparing everyone, that AI is powerful but not smart, and that AI-fueled debt and meme-stock trading signal a bubble.
12 key pointsGarfield rose from log-cabin poverty through education to the presidency, which he never wanted, and died mainly because his doctors rejected Lister's antiseptic ideas.
10 key points · 2 quotes