loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Authors: Johannes Schneider 1 ; Christian Meske 2 and Michalis Vlachos 3

Affiliations: 1 University of Liechtenstein, Vaduz, Liechtenstein ; 2 University of Bochum, Bochum, Germany ; 3 University of Lausanne, Lausanne, Switzerland

Keyword(s): Explainability, Artificial Intelligence, Deception, Detection.

Abstract: Artificial intelligence (AI) comes with great opportunities but can also pose significant risks. Automatically generated explanations for decisions can increase transparency and foster trust, especially for systems based on automated predictions by AI models. However, given, e.g., economic incentives to create dishonest AI, to what extent can we trust explanations? To address this issue, our work investigates how AI models (i.e., deep learning, and existing instruments to increase transparency regarding AI decisions) can be used to create and detect deceptive explanations. As an empirical evaluation, we focus on text classification and alter the explanations generated by GradCAM, a well-established explanation technique in neural networks. Then, we evaluate the effect of deceptive explanations on users in an experiment with 200 participants. Our findings confirm that deceptive explanations can indeed fool humans. However, one can deploy machine learning (ML) methods to detect seeming ly minor deception attempts with accuracy exceeding 80% given sufficient domain knowledge. Without domain knowledge, one can still infer inconsistencies in the explanations in an unsupervised manner, given basic knowledge of the predictive model under scrutiny. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 18.217.208.220

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Schneider, J.; Meske, C. and Vlachos, M. (2022). Deceptive AI Explanations: Creation and Detection. In Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-547-0; ISSN 2184-433X, SciTePress, pages 44-55. DOI: 10.5220/0010768300003116

@conference{icaart22,
author={Johannes Schneider. and Christian Meske. and Michalis Vlachos.},
title={Deceptive AI Explanations: Creation and Detection},
booktitle={Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2022},
pages={44-55},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010768300003116},
isbn={978-989-758-547-0},
issn={2184-433X},
}

TY - CONF

JO - Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - Deceptive AI Explanations: Creation and Detection
SN - 978-989-758-547-0
IS - 2184-433X
AU - Schneider, J.
AU - Meske, C.
AU - Vlachos, M.
PY - 2022
SP - 44
EP - 55
DO - 10.5220/0010768300003116
PB - SciTePress