The personalization–misrepresentation continuum: Reconceptualizing deception in AI-driven marketing communications

Marketing communications scholarship has treated personalization and deception as separate phenomena occupying opposite ends of an ethical spectrum. This separation has grown untenable as generative artificial intelligence enables firms to tailor communications using inferred psychological, behavioral, and biometric data that consumers neither disclosed nor anticipated. This paper proposes that personalization and misrepresentation lie on a single conceptual continuum defined by the degree to which message tailoring exploits information asymmetries between firm and consumer. Three antecedents move communications along the continuum: opacity of the personalization logic, depth of inferred psychological data, and divergence between stated consumer preferences and firm optimization objectives. Seven propositions specify the conditions under which consumers reclassify personalization as misrepresentation, including a category-specific threshold effect in which vulnerability-relevant domains such as health and finance trigger earlier reclassification, a transparency backfire in which disclosure of deep inference erodes trust even when optimization favors the consumer, and a hysteresis dynamic in which reclassification resists reversal. The framework integrates the personalization, deception, privacy, and AI persuasion literatures, is illustrated with documented industry practices, and offers scholars an operationalizable construct and regulators a defensible conceptual boundary for policy debates on AI-driven marketing communications.