The Neuron: AI Explained Building Mathematical Superintelligence: A Stanford Dropout's $64M Bet on AI Math
Dec 30, 2025
Karina Hong, a Stanford dropout and founder of Axiom Math, is on a mission to create a self-improving mathematical reasoning AI. She reveals how her team tackled a 130-year-old problem and disproved a longstanding conjecture, showcasing the power of math as a foundation for fields like coding and finance. Karina discusses the challenges of auto-formalizing proofs and highlights Axiom's vision of an IDE co-pilot that aids researchers and traders alike. The exciting intersection of neuroscience and AI is explored, promising a revolutionary leap in mathematical discovery.
AI Snips
Chapters
Transcript
Episode notes
Grid Cells Reveal Hexagonal Math
- Carina recounts studying grid cells at Oxford and linking hexagonal firing patterns to energy-minimizing tilings.
- She highlights this as a striking instance of deep math appearing in neuroscience.
What Mathematical Superintelligence Means
- Carina Hong defines mathematical superintelligence as an AI that proves theorems, conjectures, and self-improves by closing human gaps in verifying technical lemmas.
- She argues such a system can both match top mathematicians and inspire superhuman discovery at scale.
Data Scarcity And Auto-Formalization Bottlenecks
- Major bottlenecks for AI math are data scarcity and auto-formalization, since Lean proof data is tiny compared to code corpora.
- Without reliable auto-formalization, models can't scale because informal proofs in English won't convert to formal Lean code.