OpenAI adds watermarks to AI-generated images.

OpenAI has incorporated specifications from the Internet Content Validation Alliance (C2PA) into DALL-E 3. All images generated by the DALL-E 3 or ChatGPT serving APIs will include a visible watermark to identify the image as AI-generated.

OpenAI adds watermarks to AI-generated images using DALL-E 3 so users can distinguish whether the image was created by the machine.

The watermark will include details such as the image creation date and the C2PA logo in the upper left corner. This is intended to clearly indicate to users whether the image was created by a human or AI. OpenAI states that the watermark will not affect image quality or creation speed, but may increase file size by 3-5% via API and 32% when created using ChatGPT.

However, there are still ways for users to remove the AI ​​origin of an image. According to OpenAI, cropping or filtering the DALL-E output can remove this origin data.

Metadata like C2PA can easily be removed accidentally or intentionally. For example, most social media platforms today remove metadata from uploaded images, and actions like taking screenshots can also remove that metadata.

Microsoft is also applying the C2PA standard to Bing Image Creator's output images, stating: "AI-generated images from Bing Image Creator now include a digital, invisible watermark that complies with the C2PA standard."

Meanwhile, Meta has just announced that it will begin labeling content uploaded to Facebook, Instagram, and Threads if that content was created using AI. This move is part of an ongoing effort to develop industry standards for transparent labeling of AI content.

Source link: https://thanhnien.vn/openai-them-hinh-mo-vao-anh-do-ai-tao-ra-185240207164436473.htm