OpenAI details new text watermarking system for ChatGPT, Codex, and the API

OpenAI is rolling out text watermarking to help comply with the EU AI Act, with opt-in support for select API models worldwide and invisible watermarking for eligible ChatGPT and Codex output in the EU. For CIOs and technology leaders, this signals a near-term shift toward AI content provenance controls, but the technology is still imperfect: detection can produce false positives/negatives, is weakened by editing or translation, and does not prove authorship, ownership, or accuracy. IT organizations should expect new governance, compliance, and workflow requirements as AI-generated content becomes more traceable but not fully trustworthy as evidence.