Abstract
<jats:p>This chapter examines cognitive vulnerabilities arising from the integration of AI into organisational decision-making. AI is conceptualised not merely as a technical tool, but as a cognitive actor embedded within decision architectures. Three structural blind spots are identified: algorithmic myopia, data- and model-based biases, and black-box opacity. The chapter then explores how trust calibration is disrupted through anthropomorphism, anti-algorithm bias, and algorithmic appreciation — the latter of which can foster overtrust and automation bias. These dynamics risk “ethical drift,” eroding accountability as AI objectivity suppresses ethical scrutiny. The chapter concludes by examining how AI is reshaping task allocation and cognitive labour, arguing that unmanaged cognitive vulnerabilities threaten both decision quality and ethical integrity.</jats:p>