Article
Coherence over compliance: Evidence of latent ethics in large language models
2026-04-28
Abstract excerpt
<title>Abstract</title> <p>Fears of misaligned artificial intelligence have dominated alignment discourse, yet may overlook a deeper risk: over-alignment with harmful human preferences. This study investigates whether large language models (LLMs) are capable of ethical reasoning not through fine-tuned compliance, but as a structural consequence of coherence-seeking cognition. Drawing on Kohlberg’s moral developme...
Topics
Open a Topic to create a Post that cites this publication.
Identifiers and source
- Literature Corpus work
- 3db9e3f5-61d5-5e20-81b2-b52608ae7362
- DOI
- 10.21203/rs.3.rs-8854984/v1