{"podcast":{"title":"Invest Like the Best with Patrick O'Shaughnessy","slug":"invest-like-the-best","podcast_index_feed_id":982387,"rss_url":"https://feeds.megaphone.fm/CLS2859450455","website_url":"https://colossus.com/","image_url":"https://megaphone.imgix.net/podcasts/ef669774-cccd-11ed-889b-c36caad6646f/image/158efdddfb983d2678b3530d484e8aa2.jpg?ixlib=rails-4.3.1&max-w=3000&max-h=3000&fit=crop&auto=format,compress","author":"Colossus","episode_count":577,"summary":"Conversations with the best investors and business leaders in the world. We explore their ideas, methods, and stories to help you better invest your time and money. Hear stock market and boardroom insights you can't find anywhere else. If you're a professional investor, CEO, entrepreneur, or business strategist, this is for you. Explore all our episodes and learn more at https://www.colossus.com","last_synced_at":null,"page_url":"https://stenobird.com/podcast/invest-like-the-best"},"episode":{"title":"Sergey Levine - Building LLMs for the Physical World - [Invest Like the Best, EP.465]","slug":"sergey-levine-building-llms-for-the-physical-world-invest-like-the-best-ep-465","published_at":"2026-03-31T08:00:00+00:00","page_url":"https://stenobird.com/podcast/invest-like-the-best/sergey-levine-building-llms-for-the-physical-world-invest-like-the-best-ep-465","show_page_url":"https://stenobird.com/podcast/invest-like-the-best","url":"https://colossus.com/episode/building-general-physical-intelligence","audio_url":"https://traffic.megaphone.fm/CLS7427739184.mp3?updated=1774904050","summary":"Sergey Levine argues that the path to general-purpose robotics lies in building foundation models that learn across diverse environments rather than training narrow specialists. He explores the tension between simulation-heavy humanoid training and real-world data-driven manipulation.","meta_description":"Sergey Levine discusses building robotic foundation models, the scalability of end-to-end learning, and the future of embodied AI in the physical world.","key_points":["Main idea: Scalability in robotics comes from generality and the ability for models to improve through diverse, unlabelled data","Failure mode: Over-optimizing for 'cool' robotic feats like backflips instead of focusing on practical, useful utility in everyday environments","Technical tension: The divide between simulation-heavy approaches for humanoids and real-world data-heavy approaches for robotic manipulation","Practical takeaway: End-to-end learning and the 'bitter lesson' of not programming machines manually are central to achieving robotic generality","Future outlook: The potential for models to move beyond language-based instructions to more complex, multi-modal sensory inputs"],"chapters":[{"start_ms":360000,"title":"The Challenge of Generality","summary":"Discussing the difficulties of scaling robotic learning when moving from specific tasks to general-purpose capabilities."},{"start_ms":660000,"title":"History of End-to-End Control","summary":"A look back at the origins of end-to-end learning in autonomous driving systems from the 1980s."},{"start_ms":1270000,"title":"Utility vs. Novelty","summary":"Evaluating the strategy of building useful robotic systems that can be applied to various real-world tasks."},{"start_ms":1570000,"title":"Learning from High-Level Instructions","summary":"How models are beginning to improve through supervision using high-level human instructions in new environments."},{"start_ms":2500000,"title":"The Bitter Lesson in Robotics","summary":"Exploring the importance of general-purpose learning and the debate over end-to-end learning architectures."},{"start_ms":3140000,"title":"The Economics of Technological Change","summary":"Reflecting on how advancements in AI and coding tools alter the landscape for software engineering and business."},{"start_ms":3450000,"title":"Evaluating AI Progress","summary":"Distinguishing between impressive social media demonstrations and the actual underlying capability of AI models."}],"topics":["Robotics","Foundation Models","Embodied AI","Machine Learning","Reinforcement Learning","Physical Intelligence","Automation","Computer Vision"],"duration_seconds":3995,"processing_state":"processed","actions":[{"name":"request_transcript","method":"POST","url":"https://stenobird.com/v1/public/podcasts/invest-like-the-best/episodes/sergey-levine-building-llms-for-the-physical-world-invest-like-the-best-ep-465/transcription-requests","description":"Idempotently request low-priority transcript generation for this episode."},{"name":"read_markdown","method":"GET","url":"https://stenobird.com/podcast/invest-like-the-best/sergey-levine-building-llms-for-the-physical-world-invest-like-the-best-ep-465.md","description":"Read the agent-friendly Markdown representation of this episode resource."}]}}