What Just Changed in the EU?
On August 2, 2026, specific transparency obligations under Article 50 of the EU's Artificial Intelligence Act became enforceable. This marks one of the first major, practical applications of the comprehensive legislation. The new rules are designed to
ensure people know when they are dealing with AI, not a human. The goal is to combat potential manipulation and misinformation, giving users the ability to make more informed choices. These rules apply immediately to all in-scope systems, even those that were already on the market before the deadline. While some parts of the broader AI Act have different, longer implementation timelines, these transparency rules were not delayed and are now active.
The Core Rule: Disclose the Chatbot
The most direct impact for many businesses is the requirement for chatbot disclosure. Any company that provides an AI system intended to interact directly with people—such as customer service chatbots, voice assistants, or other automated agents—must ensure that users are clearly informed they are interacting with an AI. The law specifies that this disclosure must be made unless it is already obvious to a reasonably well-informed person. However, official guidance suggests this exception is narrow, meaning that in most cases, an explicit notice will be required. Simply putting a note in the terms and conditions is not considered sufficient; the disclosure must be clear, accessible, and provided at the beginning of the interaction.
Beyond Bots: Deepfakes and AI Content
The transparency rules extend beyond just chatbots. They also address the proliferation of AI-generated content. The law creates obligations for both 'providers' (who build the AI) and 'deployers' (who use it). Providers of generative AI systems must ensure their outputs, like images or audio, are marked in a machine-readable format (such as a watermark) to indicate they are artificial. Deployers have a separate duty: if they use AI to create a 'deepfake'—synthetic content that looks real—they must clearly label it as such for the human viewer. The same applies to AI-generated text on matters of public interest, unless it has undergone a thorough human review process where a person takes editorial responsibility.
Global Reach and Local Impact
A crucial aspect of the AI Act is its global reach. The rules apply not only to companies based in the EU but to any organization whose AI systems are placed on the EU market or whose AI-generated outputs are used within the Union. This means a tech company in India or the United States serving customers in Germany or France must comply with these disclosure requirements. This 'extraterritorial' effect is similar to the GDPR data privacy law and positions the EU as a de facto global regulator for AI safety and transparency. Businesses worldwide now need to audit their AI tools to determine if they fall under the Act's scope.
The High Cost of Non-Compliance
The EU has backed these new rules with significant financial penalties. Failure to comply with the Article 50 transparency obligations can result in fines of up to €15 million or 3% of a company’s total worldwide annual turnover from the preceding financial year, whichever is higher. This substantial penalty is designed to ensure that even the largest technology companies take the obligations seriously. The EU's enforcement body, the AI Office, has also rolled out tools for individuals and whistleblowers to report suspected infringements, signaling a proactive approach to enforcement from day one.
