Gabriel Argues Model Quality Is Unmeasurable in Expert Domains
Gabriel argues that benchmarks cannot measure model quality in domains where users are true experts, noting math and other verifiable fields are exceptions, while business decisions and aesthetic judgments remain hard to evaluate.
2026-11-19 ~ 2026-11-19 · 2 related posts
- Gabriel: benchmarks are broken — in the domains you know best, model quality is literally unmeasurable — gabriel1 · 2026-11-19
- Gabriel's addendum: math is verifiable, but business and aesthetics stay unmeasurable — gabriel1 · 2026-11-19