Towards Ethical AI: Empirically Investigating Dimensions of AI Ethics, Trust Repair, and Performance in Human-AI Teaming

Abstract

Objective Determining the efficacy of two trust repair strategies (apology and denial) for trust violations of an ethical nature by an autonomous teammate. Background While ethics in human-AI interaction is extensively studied, little research has investigated how decisions with ethical implications impact trust and performance within human-AI teams and their subsequent repair. Method Forty teams of two participants and one autonomous teammate completed three team missions within a synthetic task environment. The autonomous teammate made an ethical or unethical action during each mission, followed by an apology or denial. Measures of individual team trust, autonomous teammate trust, human teammate trust, perceived autonomous teammate ethicality, and team performance were taken. Results Teams with unethical autonomous teammates had significantly lower trust in the team and trust in the autonomous teammate. Unethical autonomous teammates were also perceived as substantially more unethical. Neither trust repair strategy effectively restored trust after an ethical violation, and autonomous teammate ethicality was not related to the team score, but unethical autonomous teammates did have shorter times. Conclusion Ethical violations significantly harm trust in the overall team and autonomous teammate but do not negatively impact team score. However, current trust repair strategies like apologies and denials appear ineffective in restoring trust after this type of violation. Application This research highlights the need to develop trust repair strategies specific to human-AI teams and trust violations of an ethical nature.

Human-AI collaboration by design

“…Trust is vital to ensure safety and reliability, particularly in high-stakes environments where AI decisions can have significant consequences. It fosters user adoption and engagement, as people are more likely to use and benefit from AI they trust (Schelble et al, 2022). Additionally, trust is essential for ethical decision-making and effective collaboration, as it builds confidence in AI's ability to handle tasks, learn from interactions, and respect privacy and fairness.…”

Exploring Trust in Human–AI Collaboration in the Context of Multiplayer Online Games

“…With the advancement of technologies, human-AI collaboration has attracted the attention of scholars and practitioners in improving teamwork [9,12]. Team collaboration can benefit from present technological team support by adding value to teams [12].…”

“…For example, Seeber et al [12] discuss the future design strategies for human-AI team collaboration in various areas. Schelble et al [9] explore dimensions of ethics in human-AI collaboration and try to explore the effect of trust-repair strategies on trust and team performance. Hauptman et al [2] propose design suggestions for the progress of adaptive autonomous teammates.…”

“…Trust is at the core of understanding new technology adoption [7] and virtual team collaboration [8]. Trust in the team and trust in AI teammates are influenced by the interactions between humans and AI teammates [9]. In the context of human-AI collaboration, humans and AI cooperate toward the same goal and the collaboration involves the interaction between humans and AI, personal characteristics, and environmental characteristics [10,11].…”

Development of a Real-Time Trust/Distrust Metric Using Interactive Hybrid Cognitive Task Analysis

“…Trust between humans and autonomy is defined by Lee and See (2004) as "the attitude that an agent will help achieve an individual's goals in a situation characterized by uncertainty and vulnerability" (p. 54). In a HAT, trust must not only be captured by the relationships between humans and autonomy but also between human teammates and autonomous teammates (Schelble et al, 2022).…”