Understanding manipulative AI techniques and exploitative vulnerabilities

The Future of Privacy Forum (FPF) has published the second in a series of articles exploring the prohibited practices set out in the EU Artificial Intelligence Act (AI Act). This latest article focuses on Article 5, which bans AI systems that engage in harmful manipulation, deception, or the exploitation of specific human vulnerabilities. The primary objective of these provisions is to ensure that people remain capable of making independent decisions.

The article addresses two specific prohibitions under Article 5(1). The first prevents the deployment of subliminal or purposefully deceptive techniques intended to materially distort behaviour in a harmful manner. The second prohibits the exploitation of vulnerabilities related to age, disability, or socio-economic status. Together, these measures support the broader goal of promoting human-centric, trustworthy AI that respects health, safety, and fundamental rights. The FPF notes that while the AI Act introduces these specific bans, manipulative practices involving personal data or online platforms are already governed by the EU General Data Protection Regulation (GDPR) and the Digital Services Act (DSA).

You can read the first episode in the series here. 

Training Announcement: The BCS Foundation Certificate in AI examines the challenges and risks associated with AI projects, such as those related to privacy, transparency and potential biases in algorithms that could lead to unintended consequences. Explore the role of data, effective risk management strategies, compliance requirements, and ongoing governance of the AI lifecycle and become a certified AI Governance professional. Find out more.