Home Technology OpenAI Boosts DALL-E: Image Editing in ChatGPT and Style Suggestions

OpenAI Boosts DALL-E: Image Editing in ChatGPT and Style Suggestions

by Andrew

OpenAI’s DALL-E is about to receive some updates that will further help shape AI-generated images, effectively simplifying the process. The main novelty is the editing tools inserted directly into ChatGPT, both in the web and mobile versions, while using DALL-E. This means that very soon it will be possible to edit our creations with AI and without having to leave the chat. But that’s not all, as DALL-E now offers style presets to inspire in image creation, similar to Android’s AI-generated wallpaper suggestions.

OpenAI’s generative tool is currently in its third generation, and recent updates like this one seem aimed at making it easier to use. In fact, DALL-E 3 has introduced integration with ChatGPT, which frees you from the burden of having to write perfect descriptions for images. Just tell ChatGPT what you want it to create, and it will take care of the rest.

Today’s update is an extension of this feature, allowing less experienced DALL-E users to edit and refine their results directly from the ChatGPT interface.

As initially anticipated, there are finally new presets that work as a starting point and include styles such as woodcut, gothic, synthwave and hand drawing, with examples that appear when you hover your mouse over them. By choosing one, you’ll automatically add it to your description. At this point, it’s up to you to get creative.

Recently, OpenAI has also committed to addressing a slightly more pressing concern related to image creation, namely the need to make this content recognizable. Earlier this year, DALL-E 3 began adding watermarks and visible metadata to mark AI-generated images. Definitely a good start, but both elements can be easily removed by those who don’t have good intentions. This isn’t just a problem for OpenAI, but it’s certainly an important one.

Recently, OpenAI began granting limited access to a text-to-speech generation platform called Voice Engine, which can create a synthetic voice based on a clip as small as 15 seconds. The AI-generated voice can read text aloud on command, either in the same language as speech or in several other languages. In addition, the company continues to work on the development of Sora, and there are rumors that the company has already shown its tool to some Hollywood directors and studios. The presentation would have taken place privately, involving names that have not been disclosed at the moment.

You may also like

Habelas.com is your go-to hub for all things tech and computing. Dive into the latest trends, explore practical guides, and discover detailed reviews to navigate the ever-evolving digital landscape. Stay informed with our in-depth insights and make the most of cutting-edge technological innovations

All Right Reserved –  Habelas.com