Apparently, (LLMs feel or simulate anxiety)[https://www.nytimes.com/2026/02/12/opinion/artificial-intelligence-anthropic-amodei.html]. Do they also take offense at things? If we can find an 'offense' direction, do LLMs take offense at something that the average English reader would be surprised by? Key question here is: if a probe shows something you don't expect for which you don't have groundtruth, can you ever really trust it?
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{holtzman-llm-offense-2026,
author = {Holtzman, Ari},
title = {LLM Offense?},
year = {2026},
url = {https://hypogenic.ai/ideahub/idea/OSdZoruxo0jdeAas3ndI}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!