
Designing How AI Grows — Tom McGrath
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?
Publisher
Machine Learning Street Talk (MLST)
Link
podcasters.spotify.com/pod/show/machinelearningstreettalk/episodes/Designing-How-AI-Grows--Tom-McGrath-e3o8p04
Published
September 2, 2026
Duration
1h 40m
Format
Podcast
Category
Issues · Tech News
Source
RSS feed
Machine Learning Street Talk (MLST)
@mlstreettalk
#designing #aigrows #tommcgrath
