Future research agenda concerning the goals and objectives of AI system explanations for end users
| Goal/objective | Future research direction | Source |
|---|---|---|
| Understandability | Investigate different groups of end users and their ability to understand AI system explanations | Cheng et al. (2019) |
| Elucidate and determine the desired levels of understanding of AI systems of different stakeholders | This study | |
| Trustworthiness | Investigate if involving humans in the loop of AI decision making can increase trust in AI systems | Cheng et al. (2019) |
| Investigate emotional and cognitive factors involved in trust such as surprise, confusion and cognitive dissonance | Yin et al. (2019) | |
| Determine how various explanation types influence the resulting trust toward the AI system | This study | |
| Transparency | Investigate the link between transparency and trust | Eiband et al. (2018) |
| Investigate how information disclosure and presentation are linked to end users' perceived transparency of the explanations | This study | |
| Controllability | Connect the goal of controllability to understandability, transparency, trustworthiness and fairness of the system | This study |
| Fairness | Approach the issue from various psychology of justice theories such as interactional justice | Binns et al. (2018) |
| Studies regarding the fairness of AI system explanations could focus on how well end users understand the real behavior of the system based on provided explanations | This study |
| Goal/objective | Future research direction | Source |
|---|---|---|
| Understandability | Investigate different groups of end users and their ability to understand AI system explanations | |
| Elucidate and determine the desired levels of understanding of AI systems of different stakeholders | This study | |
| Trustworthiness | Investigate if involving humans in the loop of AI decision making can increase trust in AI systems | |
| Investigate emotional and cognitive factors involved in trust such as surprise, confusion and cognitive dissonance | ||
| Determine how various explanation types influence the resulting trust toward the AI system | This study | |
| Transparency | Investigate the link between transparency and trust | |
| Investigate how information disclosure and presentation are linked to end users' perceived transparency of the explanations | This study | |
| Controllability | Connect the goal of controllability to understandability, transparency, trustworthiness and fairness of the system | This study |
| Fairness | Approach the issue from various psychology of justice theories such as interactional justice | |
| Studies regarding the fairness of AI system explanations could focus on how well end users understand the real behavior of the system based on provided explanations | This study |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.