Research Interpretation of Article 14 of the EU AI Act: Human in Command in Manufacturing
摘要
The rapid integration of artificial intelligence (AI) into various sectors necessitates rigorous frameworks to ensure that human oversight remains central, aligning with regulatory requirements such as Article 14 of the European Union’s AI Act. This paper introduces the “Human in Command” model, a novel approach designed to foster effective human-AI collaboration, particularly in manufacturing environments. Our model is bifurcated into two primary components: Design4Command and FeelingInCommand. Design4Command outlines a comprehensive set of methodologies for constructing human-AI-teams that emphasize human leadership and decision-making authority, ensuring that AI systems enhance rather than replace human capabilities. FeelingInCommand, on the other hand, presents innovative metrics and tools for assessing the psychological and ergonomic factors impacting humans’ perception of control and efficacy when interacting with AI systems. Using this model, organizations can optimize the efficiency and satisfaction of human operators in AI-enhanced settings. This paper details the theoretical underpinnings of our approach, the design principles of our methodologies, and potential applications across various domains, offering a guideline for developers and policymakers aiming to preserve human oversight in the age of AI.