Lying Style

2

If you ask a model to lie about something and you can get it to (e.g. via jailbreaking, roleplaying, etc.) is that distribution noticeably different from how the model normally writes?

llms stylometry lies Implemented:https://github.com/Hypogenic-AI/lying-style-nlp-0448-claude Implemented:https://github.com/Hypogenic-AI/lying-style-nlp-92c9-codex Implemented:https://github.com/Hypogenic-AI/lying-style-nlp-9950-gemini

Chat

If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:

@misc{holtzman-lying-style-2026,
  author = {Holtzman, Ari},
  title = {Lying Style},
  year = {2026},
  url = {https://hypogenic.ai/ideahub/idea/4QKMaU77er8GPrZIYA7S}
}

Comments (0)

Please sign in to comment on this idea.

No comments yet. Be the first to share your thoughts!