Explanation Hacking: The perils of algorithmic recourse

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sullivan, Emily, Kasirzadeh, Atoosa
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916290294185984
author Sullivan, Emily
Kasirzadeh, Atoosa
author_facet Sullivan, Emily
Kasirzadeh, Atoosa
contents We argue that the trend toward providing users with feasible and actionable explanations of AI decisions, known as recourse explanations, comes with ethical downsides. Specifically, we argue that recourse explanations face several conceptual pitfalls and can lead to problematic explanation hacking, which undermines their ethical status. As an alternative, we advocate that explanations of AI decisions should aim at understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2406_11843
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Explanation Hacking: The perils of algorithmic recourse
Sullivan, Emily
Kasirzadeh, Atoosa
Computers and Society
Artificial Intelligence
We argue that the trend toward providing users with feasible and actionable explanations of AI decisions, known as recourse explanations, comes with ethical downsides. Specifically, we argue that recourse explanations face several conceptual pitfalls and can lead to problematic explanation hacking, which undermines their ethical status. As an alternative, we advocate that explanations of AI decisions should aim at understanding.
title Explanation Hacking: The perils of algorithmic recourse
topic Computers and Society
Artificial Intelligence
url https://arxiv.org/abs/2406.11843