New podcast from @PeterDiamandis Moonshots With Peter Diamandis AI Insiders Reveal Elon Musk's Master Plan to Win AI w/ AWG & Dave Blundin | EP #192 Racing giants and scrappy startups collide as AI expands from labs into everyday life, with Elon Musk’s Colossus saga taking center stage and XAI’s Grock Code Fast1 promising a new speed for coding agents. Colossus 2 is planned as a one-gigawatt data center, following Colossus 1 in 122 days, a display of brute‑force hardware scaling that dwarfs previous assumptions. The conversation highlights data centers leapfrogging one another, tools demonetizing and democratizing at an astonishing pace, and moonshot thinking reshaping what students should study. Discussion moves to education and entrepreneurship as the future career path, with MIT’s new 6E entrepreneurship track and advanced algorithms being taught to high schoolers. The panel argues the most important trend is short AI timelines; students should pursue problems they care about and apply AI to them. They reference Mark Twain about the two important days in life, then contrast the idea of a future where education evolves to create problem-solvers who can embed intelligence into practical ventures. From hardware to software, the crew analyzes the bitter lesson: scaling large data, compute, and off‑the‑shelf algorithms matter more than bespoke AI breakthroughs. They compare training versus inference, noting the geographic split: training centers still US‑based, inference centers worldwide. They discuss energy constraints, on-site generation for Colossus, and the coming reality of tiling the globe with inference compute, all while Elon’s relentless drive aims to outpace rivals with massive investments and risk capital. Grock Code Fast1’s pricing showcases a race to the bottom that isn’t really a race; demand for cheap, fast code remains inelastic, and distribution channels—coded environments versus browsers—shape access. The team ponders Meta’s engineering talent moves, mentions sustainable abundance as a framework, and considers a possible governance debate about AI-powered policy, acknowledging that many expect humans to merge with AI rather to let AI govern directly. On the technology frontier, the talks pivot to Gemini 3, Nano Banana’s image work, real-time API for versatile voice interactions, and streaming interactive models that could redefine customer service, design, and language learning. Robotics loom large: Nvidia Jetson Thor at the edge, fleet learning for mass automation, and a vision of millions of humanoid robots by 2040. Health, longevity, and biotech—psilocybin aging studies and stem‑cell reeducation—signal AI’s reach into medicine and the human lifespan, closing with an optimistic sense that sustainable abundance—digitized, democratized, and accelerated intelligence—will accompany humanity forward. transcripted.ai/episode/aiinsi…