OpenAI has enhanced ChatGPT’s image generation capabilities with DALL-E by introducing a feature that allows users to edit AI-generated images directly within the chat interface. The tech giant announced this update on X, revealing that the GPT-4 powered ChatGPT now includes a paintbrush-style tool for users to modify or add elements to generated images.

A demonstration video shared by OpenAI illustrates this new feature. It shows a user creating an image of a dog adorned with a conical party hat. The video then guides viewers through the process of expanding the image and entering an editor view. In this view, the user selects a paintbrush icon, applies it to the dog’s head, and inputs a new command, “add bows.” This action results in a refined image where red bows appear on the highlighted portions of the dog, seamlessly integrating with the original picture.

Additionally, OpenAI introduced a style inspiration feature in DALL-E 3 GPT, enabling users to choose from various pre-set styles such as woodcut, close-up, low angle, artificial lighting, and hand-drawn. This development opens up new avenues for users to explore different artistic styles and effects.

The new feature, accessible through both web and mobile applications on iOS and Android, introduces a level of interaction previously unseen in AI-driven image generation. However, it’s important to note that this advanced functionality is exclusive to subscribers of ChatGPT Plus.

This latest update from OpenAI signifies a major step forward in the realm of AI-generated imagery, offering users enhanced control and creative freedom over the images they create with ChatGPT.