Explained: Digital watermarking on AI-generated texts, photos

ChatGPT-maker OpenAI has developed a tool to detect whether its AI chatbot has been used to write essays or research papers or generate photos. However, the company is debating whether to release it publicly due to the potential complexities and broader implications of such a tool. But what is this watermarking that not only OpenAI but companies like Meta, Microsoft, Google, Adobe and others use, and why does it matter?