Guardado en:
Detalles Bibliográficos
Autores principales: Fresz, Benjamin, Göbels, Vincent Philipp, Omri, Safa, Brajovic, Danilo, Aichele, Andreas, Kutz, Janika, Neuhüttler, Jens, Huber, Marco F.
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2408.02379
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929449575907328
author Fresz, Benjamin
Göbels, Vincent Philipp
Omri, Safa
Brajovic, Danilo
Aichele, Andreas
Kutz, Janika
Neuhüttler, Jens
Huber, Marco F.
author_facet Fresz, Benjamin
Göbels, Vincent Philipp
Omri, Safa
Brajovic, Danilo
Aichele, Andreas
Kutz, Janika
Neuhüttler, Jens
Huber, Marco F.
contents Developing and certifying safe - or so-called trustworthy - AI has become an increasingly salient issue, especially in light of upcoming regulation such as the EU AI Act. In this context, the black-box nature of machine learning models limits the use of conventional avenues of approach towards certifying complex technical systems. As a potential solution, methods to give insights into this black-box - devised in the field of eXplainable AI (XAI) - could be used. In this study, the potential and shortcomings of such methods for the purpose of safe AI development and certification are discussed in 15 qualitative interviews with experts out of the areas of (X)AI and certification. We find that XAI methods can be a helpful asset for safe AI development, as they can show biases and failures of ML-models, but since certification relies on comprehensive and correct information about technical systems, their impact is expected to be limited.
format Preprint
id arxiv_https___arxiv_org_abs_2408_02379
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Contribution of XAI for the Safe Development and Certification of AI: An Expert-Based Analysis
Fresz, Benjamin
Göbels, Vincent Philipp
Omri, Safa
Brajovic, Danilo
Aichele, Andreas
Kutz, Janika
Neuhüttler, Jens
Huber, Marco F.
Computers and Society
Artificial Intelligence
Developing and certifying safe - or so-called trustworthy - AI has become an increasingly salient issue, especially in light of upcoming regulation such as the EU AI Act. In this context, the black-box nature of machine learning models limits the use of conventional avenues of approach towards certifying complex technical systems. As a potential solution, methods to give insights into this black-box - devised in the field of eXplainable AI (XAI) - could be used. In this study, the potential and shortcomings of such methods for the purpose of safe AI development and certification are discussed in 15 qualitative interviews with experts out of the areas of (X)AI and certification. We find that XAI methods can be a helpful asset for safe AI development, as they can show biases and failures of ML-models, but since certification relies on comprehensive and correct information about technical systems, their impact is expected to be limited.
title The Contribution of XAI for the Safe Development and Certification of AI: An Expert-Based Analysis
topic Computers and Society
Artificial Intelligence
url https://arxiv.org/abs/2408.02379