ChatGPT's new image generator might actually be good. It's definitely easier to use

OpenAI has significantly improved ChatGPT by adding advanced image-generation capabilities utilizing its new GPT-4o model. The new release allows users to create and edit images directly within the ChatGPT platform, a radical enhancement of the website’s functionality.

The GPT-4o model now integrates the ability to generate and edit images and text. This enables ChatGPT to create accurate, realistic images, including detailed components like readable text and corporate logos. The model’s autoregressive aspect contributes to improved attribute binding and text rendering, enhancing the quality and consistency of generated visuals.

Because image generation is now native to GPT?4o, you can refine images through natural conversation. GPT?4o can build upon images and text in chat context, ensuring consistency throughout. For example, if you’re designing a video game character, the character’s appearance remains coherent across multiple iterations as you refine and experiment.

For now, it’s only available to ChatGPT Pro users

The image generation feature is already available to ChatGPT Pro users and will soon be rolled out to Plus and free users. This phased rollout enables a greater number of users to benefit from GPT-4o’s enhanced capabilities.

4o image generation rolls out starting today to Plus, Pro, Team, and Free users as the default image generator in ChatGPT, with access coming soon to Enterprise and Edu. It’s also available to use in Sora. For those who hold a special place in their hearts for DALL·E, it can still be accessed through a dedicated DALL·E GPT.

OpenAI trained GPT-4o on publicly available and proprietary data from partnerships with companies like Shutterstock. The company encourages compliance with respecting artists’ rights through policies that prevent images from being directly generated in the style of working artists. OpenAI also offers an opt-out feature for creators who do not wish their creations to be used in training sets.

Users retain sole ownership of the images generated within ChatGPT, pending the terms of service use. OpenAI has implemented robust protections to prevent abuse, such as digital watermarks indicating that generated images come from AI. These kinds of controls aim to promote responsible usage and maintain the integrity of content produced on the platform.