Legal and Economic Challenges of Automated Decision-Making by Artificial Intelligence Systems

Abstract

The widespread use of artificial intelligence systems in digital advertising and automated decision-making is reshaping both market mechanisms and traditional legal categories. This article asks to what extent Moroccan and European legal frameworks can regulate algorithmic influence, preserve consumer autonomy and allocate liability when a decision emerges from an opaque and distributed technical chain. Its objective is twofold: first, to identify the limits of legal rules based on consent, fault and fair information; second, to formulate legal and economic mechanisms capable of reconciling rights protection, digital trust and innovation. The methodology is based on a critical and conceptual literature review combining legal scholarship, behavioural economics and an analysis of the main rules applicable in Morocco and the European Union. The theoretical framework draws on information asymmetry, behavioural economics, risk-based liability and accountability. The review reveals convergence on the need for greater transparency, traceability and meaningful human oversight, while disagreements persist regarding the scope of strict liability and the role of voluntary technical standards. Economically, algorithmic targeting can lower customer-acquisition costs, improve matching and increase conversion, but it may also raise information costs, intensify consumer dependency, facilitate price or content discrimination and weaken trust. The main theoretical gap lies in the absence, in Moroccan law, of a cross-cutting governance model combining algorithmic audits, risk management, certification, technical standardisation, supervisory control and adapted evidentiary mechanisms. The article therefore proposes a risk-based framework combining binding legal duties, technical standards and economic incentives for compliance, with stronger powers for supervisory authorities and enhanced remedies for individuals affected by automated decisions.

JEL Classification: K12, K13, K20, M37, O33.

Paper type: Theoretical Research.