Morning Edition · Tuesday, August 4, 2026Published at 2:03 AM EDT · New York

AI Evaluation Turns Into a Product Market as Researchers Show Cheap Open Models Can Grade Proofs

One forecast puts model evaluation and benchmarking tools at roughly $9.6 billion by 2035, up from about $1.15 billion last year. New research tests whether small open-weight judges can replace expensive frontier graders.

The tooling layer around model evaluation is being priced as an industry rather than treated as a research chore. Precedence Research values model evaluation and benchmarking tools at about $1.15 billion in 2025 and projects roughly $9.57 b…