lesswrong8.0 / 10
Shallow Review of Technical AI Safety, 2025
Arb Research (Gavin Leech et al.)
Abstract
Third annual comprehensive review of technical AI safety research. Processed every arXiv alignment paper, all Alignment Forum posts, and a year of Twitter. Structured as ~800 links across agendas, teams, critiques, and papers. Key community orientation and field-mapping resource.
Research area
agent foundationsalignment
Published
2025
Source
lesswrong
Org
Alignment Forum
Sign in to read and join the discussion.