Assessing the adequacy of the EU Trade Secrets Directive in protecting AI-Driven Business Algorithms.

Abstract

This thesis assesses whether the EU Trade Secrets Directive adequately protects AI-driven business algorithms. It argues that trade secret law was designed for static information and struggles to accommodate self-learning, constantly evolving AI systems. The difficulty of identifying, proving, and enforcing misappropriation of dynamic algorithms significantly weakens legal protection in practice. The analysis further explores the growing tension between trade secret protection and emerging EU legislation, notably the AI Act and the Data Act, which prioritize transparency and access to information. These overlapping regimes create legal uncertainty for companies relying on proprietary AI models. The thesis concludes that without clearer rules and a balanced legal framework, innovation incentives in the EU risk being undermined.