Operationalising the definition of general-purpose AI: assessing four approaches

Abstract

The European Union’s Artificial Intelligence (AI) Act is set to be a landmark legal instrument for regulating AI technology. While stakeholders have primarily focused on fixed-purpose AI applications (also known as narrow AI), more attention is required to understand the nature and regulatory implications of highly and broadly capable systems. During the AI Act policymaking process, several definitions for general-purpose AI (GPAI) were considered, each attempting to distinguish between systems with and without a fixed purpose. This distinction is crucial, as it influences the allocation of regulatory obligations. In this article, we operationalise these differences through the concept of ‘distinct tasks’ and examine four approaches (quantity, performance, adaptability, and emergence) to determine whether an AI system should be classified as a GPAI. We suggest that EU stakeholders use the four approaches as a starting point to discriminate between fixed-purpose AI and GPAI.