Benchmarking AI Models

Benchmarking AI Models

From Linear Digressions by Katie Malone

March 30, 2026 · 30 min

About this episode

This episode explores how to benchmark AI models and the complexities involved in comparing their performance.

How do you know if a new AI model is actually better than the last one? It turns out answering that question is a lot messier than it sounds. This week we dig into the world of LLM benchmarks — the standardized tests used to compare models — exploring two canonical examples: MMLU, a 14,000-question multiple choice gauntlet spanning medicine, law, and philosophy, and SWE-bench, which throws real GitHub bugs at models to see if they can fix them. Along the way: Goodhart's Law, data contamination, canary strings, and why acing a test isn't always the same as being smart.

People in this episode

Host: Katie Malone

Topics covered

  • AI models
  • benchmarking
  • LLM benchmarks
  • Goodhart's Law
  • data contamination
  • canary strings

Keywords

  • AI models
  • benchmarking
  • MMLU
  • SWE-bench
  • Goodhart's Law
  • data contamination
  • canary strings

Mentioned in this episode

Organizations: MMLU, SWE-bench, GitHub

More episodes of Linear Digressions

Explore listener stats, chart rankings, contacts and more on the Linear Digressions podcast page.