OpenAI Watermarks ChatGPT Outputs by Default in EU

OpenAI has announced plans to enable digital watermarking for all ChatGPT text outputs within the European Union. The initiative, which will activate by default for regional users, seeks to improve content transparency and support efforts to identify AI-generated material. Industry observers note, however, that the underlying technology carries inherent limitations shared by existing synthetic media detection systems. The embedded signals remain relatively unreliable and can be easily circumvented through minor text alterations. This policy adjustment aligns with broader European regulatory expectations for generative AI accountability, even as technical constraints highlight the ongoing difficulty of achieving foolproof content tracing. OpenAI’s regional rollout establishes a clear benchmark for proactive labeling practices, though the measure’s practical impact will be moderated by the known vulnerabilities of current watermarking frameworks.