Exploring User Trust in AI-Generated Summaries: A Human-Computer Interaction Perspective

چکیده

User trust in AI-generated summaries is a pivotal factor influencing the adoption and efficacy of artificial intelligence tools in various domains. This paper explores the nuanced dimensions of trust from a Human-Computer Interaction (HCI) perspective, aiming to elucidate the mechanisms by which users engage with, and develop confidence in, AI-driven summarization technologies. The study employs a mixed-methods approach, integrating quantitative analysis of user interaction data with qualitative assessments gathered through interviews and surveys. This dual framework facilitates a comprehensive understanding of trust dynamics, encompassing both cognitive and affective components.

Our investigation highlights critical variables that impact user trust, including perceived accuracy, transparency, and the explainability of the AI's decision-making processes. We analyze how these factors interact with user-specific variables such as prior experience with AI, domain knowledge, and individual propensity to trust technology. The findings suggest that enhancing user trust requires a multifaceted strategy that combines technological improvements with user-centric design principles. Moreover, the role of feedback loops in fostering a trusted interaction environment is examined, underscoring the importance of iterative user feedback in refining AI-generated outputs.

The implications of this research extend to the design and deployment of AI summarization tools, emphasizing the need for interfaces that support user understanding and engagement. Recommendations include the incorporation of user-controlled customizability features, transparent algorithmic explanations, and adaptive interfaces that respond to user feedback. By addressing these factors, developers can create more trustworthy AI systems that align with user expectations and enhance overall satisfaction.

This study contributes to the broader discourse on AI ethics and technology acceptance, offering valuable insights for researchers and practitioners aiming to foster user trust in AI applications. By bridging the gap between technical capabilities and human-centered design, the paper provides a roadmap for developing AI systems that are not only efficient but also trustworthy and user-friendly.