How to Catch an AI Liar: Lie Detection in Black-Box LLMs by Asking Unrelated Questions

Lorenzo Pacchiardi, A. J. Chan, S. Mindermann, Ilan Moscovitz, Alexa Y. Pan +3 more
2/12/2026

No official code found. Best implementation: LoryPack/LLM-LieDetector (74 stars, BSD-3-Clause, last commit 2024-06-19).

Semantic Scholar

Code Implementations(1)

LoryPack/LLM-LieDetectorCommunity63%
BSD-3-Clause

Code for the ICLR 2024 paper "How to catch an AI liar: Lie detection in black-box LLMs by asking unrelated questions"

7415Shell, PythonSep 11, 20232 years agoBSD-3-Clause

Similar Papers

Stabilizing Zero-Shot Prediction: A Novel Antidote to Forgetting in Continual Vision-Language Tasks

90% match2026

Orthogonality-Enforced Latent Space in Autoencoders: An Approach to Learning Disentangled Representations

89% match1 repo132026

Sparse Beats Dense: Rethinking Supervision in Radar-Camera Depth Completion

88% match1 repo352026

Knowledge Retention in Continual Model-Based Reinforcement Learning

88% match1 repo12026

Pseudo-Calibration: Improving Predictive Uncertainty Estimation in Unsupervised Domain Adaptation

88% match2026

MM-RLHF: The Next Step Forward in Multimodal LLM Alignment

88% match1 repo1992026

Cite this paper

@article{pacchiardi2026catch,
  title  = {How to Catch an AI Liar: Lie Detection in Black-Box LLMs by Asking Unrelated Questions},
  author = {Lorenzo Pacchiardi and A. J. Chan and S. Mindermann and Ilan Moscovitz and Alexa Y. Pan and Yarin Gal and Owain Evans and J. Brauner},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:3b1fd2a7dc1490276ee0bed4a4bfab06a205224b},
  journal = {ICLR 2024 2024}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/3b1fd2a7dc1490276ee0bed4a4bfab06a205224b.svg)](https://codeofpaper.com/paper/3b1fd2a7dc1490276ee0bed4a4bfab06a205224b)

Discussion