Terence Tao

Mathematician widely regarded as one of the greatest living mathematicians and perhaps the most productive across the broadest range of mathematical fields. Fields Medal recipient (2006), Breakthrough Prize, Royal Medal. Professor at UCLA. Often called “the Mozart of Mathematics.”

Background

Works across harmonic analysis, partial differential equations, combinatorics, number theory, analytic number theory, and mathematical physics. Known for bridging distant fields — identifying that the correct language from one area unlocks a problem in another. Self-described mathematical “fox” (knowing many things across fields) rather than “hedgehog” (knowing one field very deeply), though capable of both modes. Collaborated extensively on the Kakeya Problem and Szemerédi-type results; proved global regularity for wave maps at critical energy (2000); constructed the averaged Navier-Stokes blowup (2016). Advocate for formal proof verification (Lean) and AI-assisted mathematics.

Appearances in this wiki

Key positions

• The hardest research problems lie at the boundary between tractable and hopeless — where existing techniques cover 90% and only the last 10% requires new insight
• Strategic cheating: isolate one difficulty at a time; never fight ten opponents simultaneously
• Structure–randomness dichotomy: any mathematical object is either random (no pattern) or structured (close to a simpler algebraic object); inverse theorems make this precise
• Supercriticality distinguishes Navier-Stokes (3D, hard, possibly blows up) from its 2D version (proved regular by Ladyzhenskaya); weather prediction unpasteable past two weeks for the same reason
• A good physical theory is a compression of observations: fewer parameters than data points
• The Hamiltonian is the “right” concept connecting classical and quantum mechanics — notational convergence across fields signals that you have the correct abstraction
• Lean + Mathlib + LLM lemma search is already useful; the current bottleneck in formal proof is retrieval, not reasoning
• AI accelerates mathematical exploration (quick plots, simple iterations) by reducing the coding tax