{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Stewart Squared","title":"Episode #84: From World Models to Robot Orchestras: Inside the New Stack of Real-Time Intelligence","html":"","width":"100%","height":180,"duration":3007,"description":"This week on Stewart Squared, Stewart Alsop sits down with his father Stewart Alsop II — veteran tech journalist, former editor of InfoWorld, and longtime Silicon Valley venture capitalist — for a wide-ranging conversation that moves from the origins of the CPU and operating systems all the way to the geopolitical chip war playing out between ARM, Intel, RISC-V, and China's SMIC. Along the way they get into NVIDIA's push into CPUs, the difference between LLMs and world models, Waymo's autonomous driving stack, and what it actually feels like to orchestrate a swarm of AI coding agents while building four apps at once. Stewart II references a Ben Thompson Stratechery interview with Rene Haas, CEO of ARM, worth checking out: https://stratechery.com/2024/an-interview-with-arm-ceo-rene-haas/\nTimestamps\n00:00 — CPU history and why mainframes never had a central processing unit\n05:00 — Jensen Huang's five-layer cake and the slowdown in LLM training data\n10:00 — Ring zero, operating systems, and the shift from mainframes to personal computers\n15:00 — ARM architecture, Apple's chip transition, and the Wintel breakup\n20:00 — RISC-V as an open-source ISA and China's play for chip sovereignty\n25:00 — TSMC vs SMIC, the node gap, and Intel's foundry ambitions\n30:00 — Real-time inference vs batch LLM training and what that means for AI\n35:00 — Stewart Jr.'s coding agent setup and the chaos of managing planning agents in parallel\n40:00 — Hallucinations, probabilistic vs deterministic systems, and staying in the loop\n45:00 — Competitive landscape of LLMs and the race toward general world models\n48:00 — Fei-Fei Li's World Labs, Waymo's driver model, and the robot orchestra idea in Buenos Aires\nKey InsightsThe CPU was never part of mainframe architecture — it was a concept born with the personal computer. Once Intel and Motorola introduced the first chips, everything from operating systems to software stacks got built outward from that core, and that architecture eventually...","thumbnail_url":"https://img.transistorcdn.com/QlC48R1fevmul8S_Y7L1P2WUMnEKqwNUR7TBsNqFHaE/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS84NDMx/NWQ0MWQ1Y2NkNjhk/ZDI1NWE0MTUyN2Q5/Zjg0NS53ZWJw.webp","thumbnail_width":300,"thumbnail_height":300}