{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Crazy Wisdom","title":"Episode #520: Training Super Intelligence One Simulated Workflow at a Time","html":"","width":"100%","height":180,"duration":3004,"description":"In this episode of the Crazy Wisdom podcast, host Stewart Alsop sits down with Josh Halliday, who works on training super intelligence with frontier data at Turing. The conversation explores the fascinating world of reinforcement learning (RL) environments, synthetic data generation, and the crucial role of high-quality human expertise in AI training. Josh shares insights from his years working at Unity Technologies building simulated environments for everything from oil and gas safety scenarios to space debris detection, and discusses how the field has evolved from quantity-focused data collection to specialized, expert-verified training data that's becoming the key bottleneck in AI development. They also touch on the philosophical implications of our increasing dependence on AI technology and the emerging job market around AI training and data acquisition.Timestamps00:00 Introduction to AI and Reinforcement Learning03:12 The Evolution of AI Training Data05:59 Gaming Engines and AI Development08:51 Virtual Reality and Robotics Training11:52 The Future of Robotics and AI Collaboration14:55 Building Applications with AI Tools17:57 The Philosophical Implications of AI20:49 Real-World Workflows and RL Environments26:35 The Impact of Technology on Human Cognition28:36 Cultural Resistance to AI and Data Collection31:12 The Bottleneck of High-Quality Data in AI32:57 Philosophical Perspectives on Data35:43 The Future of AI Training and Human Collaboration39:09 The Role of Subject Matter Experts in Data Quality43:20 The Evolution of Work in the Age of AI46:48 Convergence of AI and Human ExperienceKey Insights1. Reinforcement Learning environments are sophisticated simulations that replicate real-world enterprise workflows and applications. These environments serve as training grounds for AI agents by creating detailed replicas of tools like Salesforce, complete with specific tasks and verification systems. The agent attempts tasks, receives feedback on failures, and...","thumbnail_url":"https://img.transistorcdn.com/UZbrDrlO5VTfDNcq188THwbv0T09vcmLyzx3BcPI9bs/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS81Y2Rj/OGFiMTYyMGFkNTM5/N2NjOWI2MWM5YzQ1/YTc2Ny5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}