Future research agenda concerning the current general research profile of XAI and AI explanations for end users
| Future research direction | Sources |
|---|---|
| Validate the findings in real-world scenarios | Chazette and Schneider (2020), Cirqueira et al. (2020), Cheng et al. (2019), Eiband et al. (2018), Eslami et al. (2018), Lim and Dey (2009), Ngo et al. (2020) |
| Consider XAI aimed at various stakeholders | Binns et al. (2018), Brennen (2020) |
| Clarify XAI terminology and conceptualizations | Chazette and Schneider (2020) |
| Explore interactions between AI explanations and other design aspects | Chazette and Schneider (2020) |
| Investigate how end users' focus on explanations change over time, i.e. whether they at first focus more on whether they can trust the system and later other aspects | Cramer et al. (2008) |
| Explore the impacts of allowing users to question the AI system's decisions | Ehsan et al. (2019) |
| Investigate and elucidate industry- and profession-specific explanation needs | This study |
| Investigate the role of end users' education and understanding on understanding provided explanations | This study |
| Focus on explaining the dark side and unwanted consequences of AI systems | This study |
| Future research direction | Sources |
|---|---|
| Validate the findings in real-world scenarios | |
| Consider XAI aimed at various stakeholders | |
| Clarify XAI terminology and conceptualizations | |
| Explore interactions between AI explanations and other design aspects | |
| Investigate how end users' focus on explanations change over time, i.e. whether they at first focus more on whether they can trust the system and later other aspects | |
| Explore the impacts of allowing users to question the AI system's decisions | |
| Investigate and elucidate industry- and profession-specific explanation needs | This study |
| Investigate the role of end users' education and understanding on understanding provided explanations | This study |
| Focus on explaining the dark side and unwanted consequences of AI systems | This study |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.