Skip to content
AFTER CERTAINTY

Person

Dan Hendrycks

AI safety researcher and director of the Center for AI Safety, known for benchmarking and holistic evaluation of language models.

Why this matters

Hendrycks pushes measurement before myth—when models scale faster than understanding, evaluation becomes the correction loop.

Focus in observatory

Works

article

Holistic Evaluation of Language Models.

Hendrycks, Dan, Collin Burns, Anya Chen, and Spencer Ball. "Holistic Evaluation of Language Models." *Annals of the New York Academy of Sciences* (2023). https://doi.org/10.1111/nyas.15007.

Related books

Book

Coupling

Cohesion, Consequence, and the Architecture of Responsibility

A practical framework for seeing where systems are tightly joined, where they drift apart, and how those coupling choices shape accountability over time.