A carregar...
Publicação
The Impact of Explanation Granularity on Perceived Trust in AI Systems
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.16 MB | Adobe PDF |
Orientador(es)
Resumo(s)
The rapid integration of Artificial Intelligence (AI) systems across various industries has raised
significant concerns regarding user trust, particularly in systems that handle sensitive personal data.
One of the key challenges in AI adoption is ensuring transparency and understanding, as users often
perceive these systems as opaque "black boxes." This study investigates how different levels of
explanation granularity (less detailed versus more detailed) impact users' perceptions of key attributes
such as privacy, fairness, reliability, and transparency, and how these perceptions, in turn, affect trust
in AI systems. Through an examination of the relationship between explanation granularity and user
trust, this research explores whether more detailed explanations foster greater trust in AI
technologies. The study reveals that while detailed explanations can influence users’ perceptions of
these attributes, their impact on trust is more complex, with reliability being the primary driver of
trust. The findings suggest that while detailed explanations are important for shaping perceptions, they
do not always lead to higher trust, highlighting the need for a balanced approach in designing
explainable AI systems. This research contributes valuable insights for researchers, designers, and
policymakers aiming to develop AI systems that are not only transparent but also aligned with user
expectations and ethical standards.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Business Analytics
Palavras-chave
Artificial Intelligence Explainable AI Explanation Granularity Trust Transparency Reliability Privacy Fairness SDG 9 - Industry, innovation and infrastructure SDG 16 - Peace, justice and strong institutions