San Francisco: (RightNow) OpenAI has integrated an invisible watermark in its chatbot ChatGPT and Codex-generated texts. This initiative aims to enable machines to identify AI-generated content.

The company announced on October 5 that this change will be mandatory in the European Union, while it will be optional in other regions. This modification will assist in identifying content written by large language models.

OpenAI’s new text watermarking system, ‘Text Green,’ will be used for this purpose. This system will help AI detection machines determine if a text is AI-generated.

This change will be introduced in ChatGPT and Codex in EU countries in the coming weeks. For selected model API users worldwide, this new feature will be optional.

It should be noted that this watermark tool will only be available to researchers and experts to verify AI-generated texts.