Perceived Accuracy and User Behavior: Exploring the Impact of AI-Based Air Quality Detection Application (AIKU)

Qurotul Aini,Irwan Sembiring,Adi Setiawan,Iwan Setiawan,Untung Rahardja

DOI: https://doi.org/10.30997/ijar.v4i3.356

2023-12-07

Indonesian Journal of Applied Research (IJAR)

Abstract:The accuracy of air quality detection is a crucial aspect influencing user trust and satisfaction with artificial intelligence (AI) based air quality detection applications. However, only a few studies have tested the effect of the accuracy of AI-based quality detection on users' acceptance and use of these applications. This study aims to fill this gap by addressing the impact of perceived accuracy on behavioral intention and behavior using the AIKU application. This research uses a quantitative approach with the online survey method, distributed in January 2023 - February 2023 to AIKU users. Valid data were 287 respondents from 317 who were received and analyzed using partial least squares structural equation modeling (PLS-SEM). This study uses a modified technology acceptance (TAM) model by adding perceived intelligence as a mediating variable between perceived accuracy and usefulness. The results showed that nine hypotheses were accepted from the 13 hypotheses proposed. The results section of hypothesis testing shows that the effect of perceived AIKU application accuracy on perceived usability and ease of use is insignificant. However, these influences indirectly affect the behavioral intentions and attitudes of users. Even if users do not perceive purity as an essential factor, the user's attitude towards the application is still positive. This study makes a theoretical contribution by developing the TAM model by incorporating variables of perceived accuracy and perceived intelligence relevant to the AI-based context.