Research
The Evaluation Crisis Nobody Wants to Talk About
Public benchmarks are saturated. Private evals are unreproducible. The field is flying half-blind and pretending otherwise.
June 17, 2026 · 8 min read
Section
Research
2 essays on research from The Model Memo, a free daily newsletter on AI.
Research
Post-Training Eats Pretraining
The most important capability gains of the last year didn't come from bigger models. They came from longer, weirder, more deliberate post-training.
June 12, 2026 · 9 min read
Subscribe
Get the next one in your inbox.
A free daily newsletter on AI. Always free.
Free daily newsletter. Unsubscribe anytime. No spam — ever.