Trust, Credibility and Transparency in Human-AI Interaction - Why We Need Explainable and Trustworthy AI - and Why We Need It Now

Oct 15, 2026

Uploaded by Mabel

Generative Artificial Intelligence (GenAI) has rapidly evolved to perform complex tasks across diverse domains, but its effectiveness depends on how much it is trusted by individuals, organizations, and society. The paper links trust, credibility, and transparency as interrelated requirements, emphasizing that AI “black boxes” intensify public concern due to limited explanations. Trust and credibility are multifaceted, including technical, ethical, social, and psychological dimensions. Educational adoption particularly requires strong transparency to reduce fear, improve learning outcomes, and secure a social license for AI-driven interventions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Highlights

What is already known about this topic

What this paper contributes

Implications for theory, practice and/or policy

Introduction: What Would It Be Like Without Trust and Credibility?

Why does Generative AI matter?

State of the Art in Generative AI Regarding Trust and Credibility

Trust, Credibility and Transparency in Human-AI Interaction: Why We Need Explainable and Trustworthy AI and Why We Need It Now? Aras Bozkurt, Ramesh C Sharma Abstract: Generative Artificial Intelligence (GenAI) has rapidly evolved to perform complex tasks across diverse domains. Despite its potential to redefine how we work and learn, generative AI’s effectiveness hinges on the extent to which it is trusted—by individuals, organizations, and broader societal systems. At the heart of this issue lie three interrelated concepts: trust, credibility, and transparency. In particular, the opaque nature of AI “black boxes,” where sophisticated machine learning algorithms yield outcomes without clear explanations, exacerbates public concern and highlights the necessity of more explainable, responsible AI solutions. Current literature and practice indicate that trust and credibility in AI are multifaceted, encompassing technical, ethical, social, and psychological considerations. This complexity is compounded in educational settings, where generative AI’s integration demands robust transparency to mitigate fear, enhance learning outcomes, and secure a social license for AI-driven interventions. Explainable and trustworthy AI stands out as a dynamic paradigm shift, offering interpretability at both model and outcome levels. This approach enables end-users and developers alike to examine the rationale behind AI-driven decisions, preserving human oversight and reinforcing user confidence. However, merely defining explainable and trustworthy AI does not ensure its adoption: the ongoing challenge lies in building AI systems that are simultaneously innovative, transparent, and robust. Moving forward, the credibility and long-term sustainability of AI applications will depend on our collective ability to integrate technical refinements, adaptive regulations, and societal dialogue. By doing so, we can harness GenAI’s vast potential as a transformative force—guided by enduring human values rather than overshadowed by unchecked power. Keywords: artificial intelligence, generative AI, AI in education, AIED, trust and credibility, explainable AI, reliable AI, trustworthy AI, teaching and learning, higher education. Highlights What is already known about this topic: • Trust and transparency are fundamental to AI acceptance and social license. • Generative AI’s “black box” nature heightens user concerns about risks and uncertainties. What this paper contributes: • Positions explainable and trustworthy AI as a core strategy for fostering trust and credibility. • Proposes a continuous, evolving approach emphasizing transparency, accountability, and fairness. Implications for theory, practice and/or policy: • Highlights the need for ethical, technical, and societal collaboration in AI deployment. • Calls for updated guidelines, stakeholder dialogue, and trust-driven innovation across sectors. Introduction: What Would It Be Like Without Trust and Credibility? “All the world is made of faith, and trust, and pixie dust.” ―J. M. Barrie (articulated in Peter Pan) Generative AI is a powerful and transformative technology with the potential to significantly impact diverse fields. However, its broader acceptance depends on the degree of trust it inspires, the reliability of its outputs, and the transparency of human-AI interactions (Bozkurt et al. , 2023; Bozkurt et al. , 2024) . If we can establish trust, credibility, and transparency, we can decide whether this technology will be a tool or an agent in the various ecosystems in which we interact and engage in. This being the case, it is critical to examine these concepts and justify why we need not only generative AI, but also versions that are explainable, trustworthy, responsible, and worthy of public confidence. Why does Generative AI matter? Generative AI is a sophisticated technology with immersive capabilities (Bozkurt, 2023a, 2023b, 2024; Bozkurt & Sharma, 2024; Tlili

FAQs

Why are trust, credibility, and transparency central to human-AI interaction?

They determine whether generative AI is accepted and whether people can evaluate its reliability and explanations. The paper argues that aligning these concepts supports human oversight and sustained confidence.

How does the “black box” nature of AI affect public concern?

Opaque mechanisms make decision processes hard to understand, increasing uncertainty and perceived risk. This limitation strengthens the need for explainable and responsible AI.

What role does explainable and trustworthy AI play in education?

Educational integration requires robust transparency to mitigate fear, support learning outcomes, and maintain a social license for AI-driven interventions. The paper frames explainability as a shift that enables interpretability at model and outcome levels.

Oct 15, 2026

Uploaded by Mabel