OpenAI Introduces DALL-E 3, allows artists to opt out of training

Calvin D

OpenAI has taken a significant step forward with the release of DALL-E 3, an upgraded version of its innovative text-to-image generation tool. This new iteration leverages ChatGPT, OpenAI's renowned AI chatbot, to simplify the process of generating images from textual prompts.

Artificial intelligence has made substantial progress in transforming textual descriptions into various styles of artwork, spanning from photorealistic to fantastical. However, crafting the perfect prompt has posed a challenge, even evolving into a specialized field known as "prompt engineering."

DALL-E 3, with its integration of ChatGPT, offers a solution. Subscribers to OpenAI's premium ChatGPT plans, including ChatGPT Plus and ChatGPT Enterprise, can now input their image requests and refine them through interactive conversations with the chatbot. The results are delivered seamlessly within the chat application.

ChatGPT enhances prompts, even those consisting of just a few words, by making them more descriptive, thereby offering clearer guidance to the DALL-E 3 model.

DALL-E 3 doesn't stop at ChatGPT integration; it also introduces improvements in image quality. This enhancement is particularly noticeable when dealing with longer prompts. Additionally, it addresses historical challenges faced by image-generating models, such as text and the depiction of human hands.

Furthermore, DALL-E 3 incorporates new mechanisms designed to mitigate algorithmic bias and enhance safety. For instance, it will reject requests seeking images in the style of living artists or featuring public figures. In a significant development, artists now have the option to opt out of having their artwork used for training future iterations of OpenAI's text-to-image models. This change comes as OpenAI, along with some competitors, faces legal scrutiny over the alleged use of copyrighted artwork in training generative AI image models.

The launch of DALL-E 3 occurs amid escalating competition in the generative AI domain, especially in image synthesis. Rivals like Midjourney and Stability AI are continually refining their image-generation models, intensifying the pressure on OpenAI to remain at the forefront.

OpenAI plans to introduce DALL-E 3 to premium ChatGPT users in October, with subsequent availability for research laboratories and API customers. The company has not disclosed whether or when it will release a free web-based version, similar to the previous DALL-E iterations (DALL-E 2 and the original DALL-E model).