What Just Happened in the EU?
On August 2, 2026, a critical set of transparency obligations from the EU's comprehensive AI Act officially came into effect. These rules, specifically from Article 50 of the act, aren't focused on the high-risk AI systems you might hear about in other
contexts, like medical devices or infrastructure. Instead, their goal is much more direct and impacts what nearly everyone sees online daily. The core idea is to ensure people know when they are interacting with AI or viewing content that isn't human-made. This applies to any company whose AI-driven products or services are used within the EU, making it a globally significant regulation.
The New Rules of Reality
The regulations introduce four key transparency requirements. First, if you're talking to a chatbot or an AI agent, its provider must ensure you know you're not interacting with a person. Second, providers of generative AI systems—those that create text, images, or audio—must now embed machine-readable watermarks into their output so it can be identified as synthetic. Third, any organisation using emotion recognition or biometric categorization systems must inform the people being analysed. Finally, and most critically for public discourse, there are strict rules for deepfakes and AI-generated text.
Targeting Deepfakes and Misinformation
The most scrutinized part of the new rules targets deceptive content. Any user deploying an AI system to create a 'deepfake'—a synthetic image, audio, or video that looks real and could be mistaken for authentic footage—must clearly label it as artificially generated. There's an exception for content that is obviously artistic or satirical. Furthermore, any AI-generated text published to inform the public on matters of public interest must also be labelled, unless it has undergone a substantial human review where a person takes full editorial responsibility. This is a direct attempt to preserve what one EU lawmaker called "the authenticity of facts online" and protect democratic processes from a flood of convincing but false information.
The Ripple Effect for Tech and India
These rules have a massive ripple effect beyond Europe. Major tech companies whose platforms are used in the EU must now adapt their systems or face hefty fines of up to €15 million or 3% of their global turnover. For new AI systems, the rules are immediate; for those already on the market, there is a grace period until December 2026 to comply with the watermarking requirement. This has significant implications for India's thriving IT and Business Process Outsourcing sector. Indian companies that develop, maintain, or deploy AI systems for European clients will now be considered 'providers' or 'deployers' under the EU law. They must integrate these new compliance standards—including transparency logs and bias testing—into their workflows, creating a demand for professionals with hybrid technical and regulatory skills.
A Blueprint for Global AI Governance?
The EU's AI Act is often seen as a global standard-setter, much like its GDPR data privacy law was. For India, which is currently shaping its own approach to AI governance, the EU's framework provides a valuable and detailed model. While India may opt for a different, innovation-focused path, the EU's emphasis on risk-based tiers and strong transparency obligations offers a clear playbook. As Indian firms align with EU standards to serve a major export market, it may accelerate the push for a robust domestic regulatory framework to ensure India remains a key player in the development of responsible and trusted AI globally.
