Detalhes bibliográficos
| Resumo: | Artificial intelligence (AI) is being increasingly used in daily life. Using these systems depends on end-users' trust in them, and auditing processes have been proposed to determine and communicate whether an AI system is trustworthy. However, many end-users lack the expertise to both assess the trustworthiness of AI systems and to understand the outputs of auditing. Certification labels have been proposed as a non-technical solution to communicating AI trustworthiness to end-users. This research tests the effectiveness of Trustworthy AI certification labels on end-users' trust in and behavioral intention to use (BIU) AI. In three studies, we show that using a simple certification label increases cognitive and affective trust in AI and BIU it. Moreover: cognitive trust mediates the label's effect; labels displaying Trustworthy AI principles further increase perceived trustworthiness and BIU; and exposure to labeled AI systems reduces trust in and BIU non-labeled systems. This work contributes to literature on trust and acceptance of AI and on the effectiveness of certification labels for signaling trustworthiness, and offers insights for the design of effective AI certification labels. |
|---|
| Autores principais: | Almeida, Filipa de |
|---|
| Outros Autores: | Glaum, Joana; Hildred, Kiall; Scott, Ian |
|---|
| Assunto: | artificial intelligence trustworthy AI auditing certification labels technology acceptance Applied Psychology Marketing SDG 9 - Industry, Innovation, and Infrastructure SDG 10 - Reduced Inequalities |
|---|
| Ano: | 2026 |
|---|
| País: | Portugal |
|---|
| Tipo de documento: | artigo |
|---|
| Tipo de acesso: | acesso aberto |
|---|
| Instituição associada: | Universidade Nova de Lisboa |
|---|
| Idioma: | inglês |
|---|
| Origem: | Repositório Institucional da UNL |
|---|