{"podcast":{"title":"Machine Learning Street Talk (MLST)","slug":"machine-learning-street-talk","podcast_index_feed_id":781643,"rss_url":"https://anchor.fm/s/1e4a0eac/podcast/rss","website_url":"https://podcasters.spotify.com/pod/show/machinelearningstreettalk","image_url":"https://d3t3ozftmdmh3i.cloudfront.net/staging/podcast_uploaded_nologo/4981699/4981699-1757416025703-f026fa81b6d04.jpg","author":"Machine Learning Street Talk (MLST)","episode_count":262,"summary":"Welcome! We engage in fascinating discussions with pre-eminent figures in the AI field. Our flagship show covers current affairs in AI, cognitive science, neuroscience and philosophy of mind with in-depth analysis. Our approach is unrivalled in terms of scope and rigour – we believe in intellectual diversity in AI, and we touch on all of the main ideas in the field with the hype surgically removed. MLST is run by Tim Scarfe, Ph.D (https://www.linkedin.com/in/ecsquizor/) and features regular appearances from MIT Doctor of Philosophy Keith Duggar (https://www.linkedin.com/in/dr-keith-duggar/).","last_synced_at":"2026-09-09T02:20:14.331532+00:00","page_url":"https://stenobird.com/podcast/machine-learning-street-talk"},"episode":{"title":"Designing How AI Grows — Tom McGrath","slug":"designing-how-ai-grows-tom-mcgrath","published_at":"2026-09-02T21:04:58+00:00","page_url":"https://stenobird.com/podcast/machine-learning-street-talk/designing-how-ai-grows-tom-mcgrath","show_page_url":"https://stenobird.com/podcast/machine-learning-street-talk","url":"https://podcasters.spotify.com/pod/show/machinelearningstreettalk/episodes/Designing-How-AI-Grows--Tom-McGrath-e3o8p04","audio_url":"https://anchor.fm/s/1e4a0eac/podcast/play/125116868/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net%2Fstaging%2F2026-8-2%2F431065420-44100-2-1fb5f17ee7e73.mp3","summary":"Tom McGrath is co-founder and Chief Scientist at Goodfire, and a former Google DeepMind researcher. He joins Tim Scarfe to ask what neural networks actually learn, whether their internal representations converge on structures in the world, and whether interpretability can extract new scientific knowledge rather than merely explain model outputs. Beginning with AlphaZero and learned modularity, the conversation moves into neural geometry: concept manifolds, reusable computation inside Llama, and why activation steering can fail when it pushes a model off-manifold. McGrath then makes the case for intentional design, using interpretability as part of the training loop. They examine controlled generalisation, features as rewards, predictive data debugging, and the uncomfortable fact that a model may recognise a hallucination or reward hack and still produce it. The discussion closes on grader awareness, oversight and collusion between adaptive agents, then returns to sparse autoencoders. SAEs are useful, McGrath argues, but they may fracture the higher-dimensional structures networks actually use. This episode was made with support from Goodfire. --- TIMESTAMPS: 00:00:00 Introduction: Can interpretability speed-run science? 00:02:03 The invisible grader 00:06:51 What AlphaZero learned from the world 00:12:24 Interpretability as a control loop 00:21:54 The forbidden method and safer interventions 00:37:36 Why models catch hallucinations too late 00:46:19 Debug the dataset before training 00:50:44 Why neural networks become modular 00:55:57 Finding the geometry inside a network 01:02:55 Why steering falls off the manifold 01:12:10 A reusable calculator inside Llama 01:17:19 From abstractions to goals 01:25:28 Reward hacking, oversight and collusion 01:37:23 Are sparse autoen…","meta_description":"Tom McGrath is co-founder and Chief Scientist at Goodfire, and a former Google DeepMind researcher. He joins Tim Scarfe to ask what neural networks actual…","key_points":[],"chapters":[],"topics":[],"duration_seconds":6014,"processing_state":"not_requested","actions":[{"name":"request_transcript","method":"POST","url":"https://stenobird.com/v1/public/podcasts/machine-learning-street-talk/episodes/designing-how-ai-grows-tom-mcgrath/transcription-requests","description":"Idempotently request low-priority transcript generation for this episode."},{"name":"read_markdown","method":"GET","url":"https://stenobird.com/podcast/machine-learning-street-talk/designing-how-ai-grows-tom-mcgrath.md","description":"Read the agent-friendly Markdown representation of this episode resource."}]}}