AI safety researcher

Ethan Perez

Google Scholar dblp

Last 12 months4 papers

Evaluation and benchmarks 1Defenses 1

AI safety papers from the last year

On this site: 29 papers

July 2026

A dataset of rated conceptual arguments
Jul 29Evaluation and benchmarksarXiv ↗

January 2026

The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?
Jan 30arXiv ↗
Constitutional Classifiers++: Efficient Production-Grade Defenses against Universal Jailbreaks
Jan 8DefensesarXiv ↗

November 2025

Natural Emergent Misalignment from Reward Hacking in Production RL
Nov 23arXiv ↗

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