OpenAI has unveiled its next-generation text-to-image model, DALL-E 3 to the public. The company mentions that DALL-E 3 can process even more nuanced prompts and generate extremely detailed and accurate images. It further says that DALL-E 3 doesn’t ignore your words or descriptions, unlike many other AI models, and follows your prompt to the last word. This is done to accurately translate your ideas into visual images.
DALL-E 3 Is Coming to ChatGPT Plus and Enterprise Users
The best part is that OpenAI has integrated Dall -E 3 into ChatGPT directly. What it means is that ChatGPT acts like your prompt partner who can turn your single line of ideas into a detailed paragraph. Basically, you don’t need to use ChatGPT plugins like Prompt Perfect as ChatGPT natively comes with this functionality for image generation.
OpenAI says that DALL-E 3 will be available to ChatGPT Plus users and Enterprise customers, starting this October. And as was the case with earlier models, images created using Dall -E 3 will be entirely free to use, and you will have all the permission to “reprint, sell or merchandise them“.
In terms of safety, OpenAI has done a remarkable job with its Red Teamers to tame DALL-E 3 from generating violent, hateful, or adult content. It automatically declines requests to generate images of public figures. What’s new, DALL-E 3 also declines requests to generate images based on the style of a living artist. This is a good effort by OpenAI to protect the interests of artists. In addition, artists and creators can now choose to remove their images and their body of work from future image generation models of OpenAI.
Finally, with DALL-E 3’s announcement, OpenAI said that it’s working on a tool that allows people to identify AI-generated images. We hope the new tool correctly identifies AI images, unlike Text Classifier which was released earlier.