What Just Changed in the EU?
On August 2, 2026, a key part of the EU's comprehensive AI Act—Article 50—officially came into effect. This provision mandates a new level of transparency for AI systems. In simple terms, companies that develop or use certain AI tools must now clearly
disclose when content is generated or manipulated by AI. This applies to a wide range of applications, from chatbots to so-called “deepfakes.” If you're talking to a bot, the company behind it has to make sure you know it's not a human. If an image, video, or audio file is synthetically created or altered to resemble real people or events, it must be labeled as such.
The Goal: Ending the AI 'Wild West'
This regulation is a direct response to the explosion of sophisticated generative AI tools and the risks they pose, including the spread of misinformation, fraud, and large-scale manipulation. For years, it has become increasingly difficult to distinguish between authentic, human-made content and highly realistic fakes. The EU’s move is designed to combat this deception by giving users the ability to make informed choices. The rules target four main scenarios: direct interactions with AI like chatbots; the use of systems that recognise emotions or categorise people biometrically; the creation of deepfakes; and the publication of AI-generated text on matters of public interest that has not had human editorial oversight.
How 'Source Verification' Will Work
The rules create distinct jobs for AI 'providers' (the developers) and 'deployers' (those who use the AI). Providers of generative AI systems must ensure their outputs are marked in a machine-readable format so they can be detected as artificial. This could involve technical watermarks or embedded metadata. For deployers—like a media company or a brand using AI content—the obligation is often to provide a clear, visible label for the end-user. The European Commission is backing a voluntary 'Code of Practice' which includes a set of icons that can be used to label AI content, and nearly 200 organisations had already signed on by the end of July 2026. Failure to comply isn't cheap, with potential fines reaching up to €15 million or 3% of a company's global annual turnover.
Who Is Affected? From Big Tech to Creators
These rules have a massive global footprint. The AI Act applies to any company, regardless of its headquarters, if it places an AI system on the EU market or if its AI-generated content is used within the EU. This means major US and Asian tech firms are directly impacted. But the regulations extend beyond Big Tech, affecting any business that uses AI to generate content. An employer using an AI chatbot for customer service, a marketing team creating synthetic images for a campaign, or a news outlet publishing AI-generated articles on public interest topics without human review are all subject to these rules. While content made before August 2 doesn't need to be retroactively labeled, the new era of compliance has begun. Systems already on the market have a short grace period, until December 2, 2026, to meet certain technical marking requirements.
The Ripple Effect for India
While India is developing its own approach to AI regulation, the EU's move is likely to set a global benchmark. This phenomenon, known as the “Brussels effect,” has been seen before with data privacy rules like GDPR. Global companies often find it easier to adopt the EU's strict standards across all their operations rather than creating different rules for different regions. As a result, users in India and elsewhere may soon start seeing more AI-generated content labels and disclosures. For Indian tech companies and creators with a European audience, compliance is now mandatory. This shift forces a global conversation about authenticity and trust online, pushing developers and publishers everywhere to think more critically about how they create and present digital content.
