• AI cybersecurity
• AI law
• cyber threats
• cybersecurity mitigation strategies
• data security
• high-risk AI systems
• human oversight
• layered defence
• regulatory framework.
• risk assessment
• risk management
• technical robustness

The rapid integration of artificial intelligence (AI) into various sectors has necessitated the development of robust legal frameworks to manage the associated cybersecurity risks. The European Union's Artificial Intelligence Act (AI Act) represents an important step towards establishing a comprehensive legal framework for AI, particularly in addressing the cybersecurity challenges posed by these technologies. This paper examines the cybersecurity provisions of the AI Act, focusing on identifying the challenges and proposing legal solutions for effective risk management.
Through a detailed analysis of the AI Law's cybersecurity requirements, in particular those related to high-risk AI systems, this research identifies the main gaps and limitations of the current regulatory approach. It examines the predominant cyber threats facing AI systems, and assesses the adequacy of the AI Act's risk-based approach to mitigating these risks. It also explores the legal and operational consequences of cyber-attacks on AI systems, highlighting the need for a more integrated and practical application of the Act's provisions.
The findings of this research contribute to the current debate on AI regulation by providing legal strategies and policy recommendations aimed at improving the cybersecurity framework of the AI Act. These recommendations aim to ensure that AI systems are not only compliant with legal requirements, but also resilient to the changing cyber threat landscape.