EU AI Act Transparency Rules Are Now In Effect. Was It a Missed Opportunity?
The EU AI Act’s transparency rules have entered into force, requiring providers and deployers to disclose AI-generated or manipulated content. The authors welcome requirements for metadata, watermarking, detection tools, interoperability, privacy and accessible disclosures, but argue that optional provenance information, weak obligations for open-weight models, limited oversight and unreliable detection undermine the framework. They warn that vague and broad exemptions—including personal use, minor edits and technically infeasible real-time labeling—could leave harmful viral deepfakes insufficiently regulated. Effective protection will require durable provenance, meaningful labeling, stronger enforcement and recourse for victims, while preserving safeguards against censorship and over-removal.
Estilo periodísticoInterpretativo
Strongly analytical and critical, advocating for stronger provenance, labeling, enforcement and victim recourse while assessing the regulation’s shortcomings.
Contexto políticoCentroizquierda
Supports stronger government regulation of AI-generated content, robust protections for victims and civil-society oversight, while also warning against over-removal of satire, dissent and human-rights documentation.






