“Dall-E 3 understands significantly more nuance and detail than our previous systems, allowing you to easily translate your ideas into exceptionally accurate images,” OpenAI said.
Heaptalk, Jakarta — OpenAI revealed Dall-E 3, its latest version of the AI system that creates images and art from text prompts (09/21). The previous version was released in September 2022 while the first one in January 2021 used a version of GPT-3.
The recent updates are built on the tendency to ignore words or descriptions by modern text-to-image systems, forcing users to learn prompt engineering. Working on that issue, the latest tool is claimed to be able to generate images that truly match the given prompt.
“Dall-E 3 understands significantly more nuance and detail than our previous systems, allowing you to easily translate your ideas into exceptionally accurate images,” OpenAI stated on its official page.
This tool is embedded in ChatGPT, allowing users to transform text prompts, from simple sentences to detailed paragraphs, into images they want to see. “When prompted with an idea, ChatGPT will automatically generate tailored, detailed prompts for Dall-E 3 that bring your idea to life,” said OpenAI. If the resulting image is not as desired, users can ask ChatGPT to make adjustments.
However, OpenAI designed Dall-E 3 to reject requests for creating images in the style of living artists. Apart from that, the tool allows users to not include their images in future image generation model training.
Currently, the AI visual tool is in research preview. The company said the tool will be available to ChatGPT Plus and Enterprise customers in October via the API and in OpenAI Labs later this fall. The images created with the tool will become the property of the user. This way, they do not need to ask OpenAI’s permission to reprint, sell, or merchandise them.
