Position: The Alignment Community is Unintentionally Building a Censor's Toolkit
Source
arxiv.org
Author
Sarah Ball, Phil Hackemann
Date
Why it matters
The same machinery that suppresses harmful output is general-purpose output control — worth understanding if you ship alignmentThe work of making AI systems actually pursue what their builders and users intend, rather than something subtly or dangerously different.Full definition → or moderation layers that a downstream operator could repurpose.
Terms in this piece · Glossary
alignment — The work of making AI systems actually pursue what their builders and users intend, rather than something subtly or dangerously different.
articleOpenAI-HuggingFace: A Reproduction & Lessons for Alignment TestingStewart Slocum, Malayandi Palan, Christopher Chute, Michael Kim, Benjamin Van Roy
articleAgreement Is Not Alignment: Divergent Moral Grounds in Human and LLM Ethical JudgmentsOctavian M. Machidon, Alina L. Machidon, Vojko Strahovnik, Mateja Centa Strahovnik, Jonas Miklav\v{c}i\v{c}, Marko Robnik \v{S}ikonja
articleAI Evaluation Should Work With HumansJan Kulveit, Gavin Leech, Tom\'a\v{s} Gaven\v{c}iak, Raymond Douglas