OpenAI has introduced an updated version of its DALL-E 2 image generation tool called (you guessed it!)… DALL-E 3. The big news on the new version has been integrated with ChatGPT, OpenAI’s conversational AI, to make it easier for users to describe the types of images they want to create.
Sometimes describing exactly what you want from an AI image generator can be difficult. Some people specialize in crafting detailed text prompts, known as “prompt engineering”, to get the best results. With DALL-E 3, ChatGPT can assist users by refining their text prompts to provide more detail and guidance to the tool.
DALL-E 3 is able to generate images that better match users’ text descriptions, especially when provided with rich, detailed prompts. It also shows improved abilities in drawing things like text and human hands.
Additionally, DALL-E 3 includes new safeguards. It will not produce images that duplicate copyrighted artistic works or depict public figures without permission. Artists now have an option to exclude their works from being used to train AI systems.
Other tech companies are also building image generation tools, so OpenAI is working to stay ahead with innovations like DALL-E 3.
In October, early access will be granted to ChatGPT users, with broader research and enterprise access planned later on. It remains to be seen if a free public version will be released as was done previously.