AI safety researcher

Dario Amodei

Anthropic

Google Scholar dblp Homepage

Defenses 1

Latest AI safety papers

On this site: 9 papers

February 2023

The Capacity for Moral Self-Correction in Large Language Models
Feb 15arXiv ↗

September 2022

Toy Models of Superposition
Sep 21arXiv ↗

May 2022

Scaling Laws and Interpretability of Learning from Repeated Data
May 21arXiv ↗

February 2022

Predictability and Surprise in Large Generative Models
Feb 15arXiv ↗

December 2021

A General Language Assistant as a Laboratory for Alignment
Dec 1DefensesarXiv ↗

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.