A bold push to clamp down on racist and discriminatory content generated by artificial intelligence (AI) is underway, with Executive Vice-President Virkkunen outlining the European Commission’s strategy to hold AI makers and platforms accountable. This policy focus impacts AI developers, online platforms, and users confronting hate speech, raising expectations — and likely tensions — around legal responsibility and content moderation.
Virkkunen’s detailed response addresses a written parliamentary question from Anthony Smith of The Left, probing how the EU will enforce liability mechanisms and obligations to curb AI’s role in spreading hateful material.
The reply references existing tools: the AI Act forbids AI designs that deceive or exploit human vulnerabilities to cause harm, covering discriminatory outputs too. High-risk AI must follow stringent data governance to avoid bias, and providers of general-purpose AI face transparency duties and must manage identified risks, including illegal or hateful content, reporting to the newly formed AI Office. These rules enter enforcement in August 2026. Complementing this, the Digital Services Act obliges online platforms to act swiftly to remove illegal content and demands very large platforms assess and reduce systemic risks, encompassing AI-generated hate speech.
The emphasis lies on strengthening EU oversight (via the AI Office and platform duties), increasing transparency, and imposing clear responsibilities on AI developers and platform operators. This approach prioritizes safeguarding fundamental rights and users’ mental health but entails increased regulatory burdens and compliance costs for both AI businesses and digital platforms.
Stakeholders most affected include AI providers who must ramp up risk assessment and governance, digital platforms facing stricter content policing, EU citizens targeted by hateful AI content benefitting from stronger protections, and regulators getting new enforcement roles to monitor these emerging challenges.
The Commission’s answer signals a firmer regulatory direction with key deadlines like August 2026 enforcement, indicating imminent institutional follow-up to monitor implementation and effectiveness of these overlapping legal frameworks.