OpenAI unveils game-changing image editing feature for ChatGPT

OpenAI has introduced a groundbreaking upgrade to its ChatGPT powered by GPT-4, enhancing its image generation capabilities. OpenAI announced that users can now directly interact with generated images within the chatbot interface. By clicking on the images, users enter an editing mode where they can utilise a paintbrush-style tool to highlight specific sections for adjustment or addition.

A demonstration video showcased user-generated images featuring a dog wearing a festive conical hat. After expanding the image and entering the editor view, the user employed a paintbrush icon to highlight parts of the dog’s head. Subsequently, they inputted a prompt “add bows,” resulting in the generation of a new image with red bows incorporated into the highlighted area while preserving the original composition.

Observers on various web platforms noted that the update was quietly rolled out earlier in the week, signaling OpenAI’s continuous efforts to enhance user experience.

This enhancement is made possible by integrating OpenAI’s DALL-E 3 model into ChatGPT, which was introduced in September 2023. The integration significantly broadened ChatGPT’s capabilities, allowing for more nuanced image manipulation and control.

Moreover, OpenAI introduced a new style inspiration feature within DALL-E 3 and GPT, enabling users to choose from various preset styles like woodcut, close-up, low angle, artificial lighting, and hand-drawn.

This advancement holds immense potential for various applications, including content creation, digital art generation, and graphic design. VentureBeat, for instance, utilises ChatGPT/DALL-E 3 and other AI art generators to produce compelling article header images and artwork for their platforms.

With these updates, OpenAI continues to push the boundaries of AI-driven creativity, empowering users to unleash their imagination and creativity like never before.