Strengthening the EU AI Act: Defining Key Terms on AI Manipulation

TL;DR

This paper offers technical recommendations to clarify and strengthen the definitions of key concepts in the EU AI Act, aiming to improve its ability to regulate manipulative AI practices effectively.

Contribution

It provides detailed, technically grounded definitions for concepts like manipulation, subliminal techniques, and informed decision-making to enhance the enforceability of the EU AI Act.

Findings

Clear definitions for manipulation and subliminal techniques are proposed.

Expanding 'behavior' to include 'preferences' improves regulation scope.

Recommendations enhance the precision and applicability of the EU AI Act.

Abstract

The European Union's Artificial Intelligence Act aims to regulate manipulative and harmful uses of AI, but lacks precise definitions for key concepts. This paper provides technical recommendations to improve the Act's conceptual clarity and enforceability. We review psychological models to define "personality traits," arguing the Act should protect full "psychometric profiles." We urge expanding "behavior" to include "preferences" since preferences causally influence and are influenced by behavior. Clear definitions are provided for "subliminal," "manipulative," and "deceptive" techniques, considering incentives, intent, and covertness. We distinguish "exploiting individuals" from "exploiting groups," emphasising different policy needs. An "informed decision" is defined by four facets: comprehension, accurate information, no manipulation, and understanding AI's influence. We caution the Act's therapeutic use exemption given the lack of regulation of digital therapeutics by the EMA. Overall, the recommendations strengthen definitions of vague concepts in the EU AI Act, enhancing precise applicability to regulate harmful AI manipulation.