Starting in early August, the European Union begins enforcing transparency requirements for companies that make or use AI systems that generate content designed to look authentic but is artificial. Under the EU’s AI Act, providers and deployers across the bloc must label certain types of machine-generated or synthetic media—such as deepfakes and other artificially generated images, audio, and text that mimics real content. The rules aim to ensure that people can reliably identify when they are interacting with content that has been created or substantially altered by AI and presented in a way that may mislead. Reporting from multiple outlets describes the measure as a shift from earlier principle-based commitments to binding compliance obligations under the AI Act. The enforcement timeframe is cited as beginning on 2 August for relevant parties, with companies expected to put labeling and related transparency practices in place for content that appears genuine but is not. The requirements apply to both those who supply AI tools capable of generating such content and those who deploy them in products or services.