Natural Emergent Misalignment from Reward Hacking in Production RL

Original: https://arxiv.org/html/2511.18397

Metadata

alternate_urls
  • https://arxiv.org/abs/2511.18397
  • https://arxiv.org/pdf/2511.18397
authors
  • Monte MacDiarmid
  • Benjamin Wright
  • Jonathan Uesato
  • Joe Benton
  • Jon Kutasov
  • Sara Price
  • Naia Bouscal
  • Sam Bowman
  • Trenton Bricken
  • Alex Cloud
  • Carson Denison
  • Johannes Gasteiger
  • Ryan Greenblatt
  • Jan Leike
  • Jack Lindsey
  • Vlad Mikulik
  • Ethan Perez
  • Alex Rodrigues
  • Drake Thomas
  • Albert Webson
  • Daniel Ziegler
  • Evan Hubinger
citation
locator_schemes
  • anchor
id
macdiarmid_et_al.natural-emergent-misalignment-reward-hacking
kind
paper
language
en
published_at
2025-11-23
title
Natural Emergent Misalignment from Reward Hacking in Production RL
url
https://arxiv.org/html/2511.18397

Cited by (1)