The European Union has implemented new AI transparency rules requiring companies to clearly label AI-generated content, including deepfakes, starting Sunday. The goal is to help users distinguish between real and synthetic online material, with potential fines for non-compliance.
These new EU regulations aim to combat the spread of disinformation and maintain public trust in online content by ensuring transparency about the origin of digital material, impacting how businesses develop and deploy AI technologies.
Starting Sunday, companies operating within the European Union must clearly label AI-generated content, including deepfakes, as part of the bloc's comprehensive artificial intelligence transparency rules. The objective is to ensure that European citizens can readily distinguish between authentic and synthetic online material.
The EU's sweeping AI law is being implemented in phases. The initial stage requires companies to make it evident when users are interacting with AI systems, such as chatbots, or when content like images or text has been produced by AI. This can be achieved through integrated watermarks or other markers that facilitate the detection of AI-generated material. Non-compliant firms risk substantial fines.
The guideline also mandates that deployers of AI systems must inform individuals when they are exposed to emotion recognition and biometric categorization tools, deepfakes, and text publications on matters of public interest that have not undergone human review or editorial control. An EU official highlighted that generative AI enables disinformation to be created at an unprecedented scale, tailored to specific audiences, and disseminated rapidly, making it difficult to discern real from synthetic content.
These rules specifically target content created for professional purposes, with individuals using AI solely for personal use being exempt. Text intended to inform the public on general interest issues will also require labeling if generated by AI without human editorial oversight. Existing AI systems have a deadline of December 2, 2026, to adapt to these new regulations, although exemptions are provided for artistic, creative, satirical, and fictional works.
While the EU faces criticism for increasing demands on businesses, tech giants have begun implementing their own labeling systems. TikTok has required content creators to label AI-generated content for years, and Meta has introduced an "AI Info" label on Instagram and Facebook. Google, a signatory to the EU's code of conduct on AI transparency, is collaborating on digital tagging tools. However, Google's Karen Massin cautioned that excessive and overlapping AI labels could create regulatory complexity and confuse the very users the rules aim to protect.