OpenAI is developing a tool that can detect images generated by its AI model DALL-E 3, which can create realistic and artistic images from text descriptions.

The tool is reportedly 99% accurate at identifying DALL-E 3 images, even if they have been modified or edited.

However, OpenAI has not decided when or how to release the tool to the public, citing safety and ethical concerns.

OpenAI wants to prevent harmful or misleading uses of DALL-E 3 images, such as spreading misinformation, violating privacy, or infringing intellectual property.

OpenAI also faces the challenge of defining what constitutes an AI-generated image, as some images may be a result of human-AI collaboration or creativity.

Read More Stories