{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Media and the Machine","title":"Physical AI: Could Training Data Become a $45B Market?","html":"","width":"100%","height":180,"duration":1635,"description":"My guest today is Sce Pike, VP of AI Growth and Solutions at TELUS Digital.\nShe built and sold two companies. Ernst & Young bought her digital consultancy, Citizen. ADT bought her smart-home company, IOTAS.\nSce now has her sights set on physical AI and robots.\nWe break down Google’s recent breakthrough where its new Gemini 4 Argon taught a robot to play cornhole. And it made seven shots in a row.\nBut what data makes skills like that possible? And who’s going to supply it?\nWe get specific about what AI labs want. And why robotics companies like Figure AI are getting into the data collection business.\nWe discuss why robots might need a billion training hours. Plus, her forecast of a $45 billion market by 2035. That’s just collecting data for physical AI.\nThen we explore China’s “data collection cities.” Imagine a hundred thousand  people recording their everyday lives – just to train robots.\nAnd why she thinks China (despite being behind the US in LLMs) can be #1 in Physical AI .\nPlus, a robotic centaur with chainsaw hands.\nAs always, we close talking about AI’s impact on families and jobs. Including advice for her kids. They’re artists—and they’re rebelling against her AI optimism.\nPlease enjoy my conversation with Sce Pike.\nThx!\nRob Kelly","thumbnail_url":"https://img.transistorcdn.com/HFcKrDy4G3ckU9bjuMpsJ7q5fCZAR2rBgUQcovom8lk/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9hMjUw/Mjk1Y2I5Nzg4Mzg3/YzZiMzk2YTQ4MTI3/NGE2ZS5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}