Knowledge on trial: AI, trust and the future of decision-making

Abstract

The purpose of the paper is to highlight that while they may be in a hurry to feed the growing giant of AI, they must also thoughtfully consider the consequences. The black-box nature of AI raises concerns about transparency and trust and as long as they do not fully understand how AI reaches its conclusions, questions about the reliability of AI-driven decision-making will continue. A qualitative secondary research approach, using content analysis of evidence collected from newspapers, research papers and recent books to critically compare AI-driven and traditional approaches to knowledge and decision-making. The findings reveal that AI-driven decisions cannot be trusted blindly, as they may reinforce existing vulnerabilities and discrimination in society. Therefore, AI must be used mindfully, with continuous human oversight, transparency and accountability, to reduce rather than exacerbate disorder and inequality in the information society, while building public trust instead of creating fear. The paper is original, offering a critical perspective on AI-driven decision-making, trust and the need for mindful human oversight in the information society.