Towards a Logical Analysis of Misleading and Trust Erosion

Misleading is, regrettably, an integral part of the commonsense world. Though lying, deception, and similar malignant variants of misleading have been thoroughly investigated in ethics and social psychology, there is a rather slim related literature within the logicist AI tradition. In this paper, we present foundations for a logical theory of general misleading, with an eye on its effect on trust erosion. In particular, we define a bare-bones notion of misleading and identify four dimensions along which we distinguish eighty one variants of misleading. Given this analysis, we suggest that a logical theory of misleading for trust erosion should include an account of belief, desire, intention, and causality. A logical language LM is sketched and used to represent the identified assortment of misleading scenarios.