BRUSSELS, BELGIUM / RankWire.AI / – European Union transparency obligations regarding artificial intelligence came into force across member states on August 2, 2026. Under Article 50 of the EU AI Act, chatbots, deepfakes, synthetic media, and certain public-interest texts are now regulated. The legislation distinguishes between technical marking requirements and notices visible to audiences. It does not mandate visible labels on all AI-created content; instead, each obligation depends on the specific system, the type of content, and how an organization presents it.
Entities offering interactive AI tools are required to disclose when they are communicating with artificial intelligence. This notice can be omitted if the user can unmistakably identify the system’s artificial origin. Additionally, generative AI providers must embed machine-readable signals into synthetic text, audio, images, and videos. These signals aim to facilitate detection tools in recognizing generated or modified media. When technically feasible and reliable, providers are obligated to implement effective marking methods using available technology.
A separate set of disclosure rules applies to organizations that publish synthetic content. They are required to identify deepfake images, videos, and recordings that could be mistaken for real. In cases involving public-interest AI-generated texts, organizations must also label such content. However, human oversight and editorial review can eliminate this specific labeling requirement. Furthermore, responsibility for published material must be accepted by an individual or organization before exemptions can be granted.
Clear notices focus on misleading synthetic media
The scope of Article 50 also covers emotion recognition and biometric categorization technologies that analyze individuals. Operators are required to inform affected parties when deploying these tools, with some limited legal exceptions. The law provides more flexible disclosure rules for artistic, fictional, creative, and satirical works, ensuring notices do not disrupt the normal viewing or enjoyment experience. Nevertheless, when such content contains deepfake elements, appropriate disclosure remains necessary.
European Commission guidance has been published to clarify transparency standards, definitions, and available exceptions. It also features a voluntary code for developers and organizations utilizing generative AI, allowing participants to demonstrate compliance through their marking and labeling methods. Companies not participating in the code are expected to implement other effective compliance measures. While optional icons can raise public awareness, they alone do not establish adherence to the EU AI Act.
Regulatory bodies acquire enforcement authority
Primary enforcement responsibilities rest with national market surveillance agencies in each EU member state. The European Commission oversees AI systems used by EU institutions and agencies. Violations of transparency requirements can result in fines of up to 15 million euros or 3% of a company’s global annual turnover. Smaller businesses may face reduced penalties based on the lesser of these two amounts.
A transitional period is granted for generative AI systems introduced before August 2, 2026. Such providers have until December 2, 2026, to comply with the machine-readable marking obligation. This extension applies solely to technical labeling requirements and does not delay disclosures related to interactive AI, deepfake labeling, or public-interest content rules. Content created before the law’s effective date is not subject to retroactive labeling obligations.