OpenAI's DALL-E 3 redefines prompting and enables artists to opt out of training

OpenAI introduces DALL-E 3, an upgraded version of its text-to-image tool that simplifies the prompting process with the help of ChatGPT.

DALL-E 3 translates text prompts into images, and this version boasts enhanced contextual comprehension compared to its predecessors. One of its key features is the integration with ChatGPT, enabling users to have ChatGPT generate prompts for DALL-E 3, removing the need for detailed prompts from users themselves. Users can still use their own prompts if they have specific ideas.

The integration with ChatGPT simplifies the process of generating AI art, making it more accessible to a broader audience. In a demonstration, ChatGPT helped create a logo for a mountain-based ramen restaurant by generating a paragraph-long prompt for DALL-E 3 to follow. This collaboration between the two AI models facilitates art creation, even for those who may not be skilled at crafting prompts.

DALL-E was originally released in January 2021 and gained attention for its text-to-image generation capabilities. However, it has been criticized for creating controversial content.

OpenAI has focused on enhancing safety measures in DALL-E 3 to prevent the creation of inappropriate or offensive images. The company collaborated with external red teamers and used input classifiers to train the model to ignore specific words that could lead to explicit or violent prompts.

Additionally, DALL-E 3 will not recreate images of public figures mentioned in the prompt. OpenAI also allows artists to opt their work out of future text-to-image AI models to address copyright concerns and potential lawsuits. Artists can submit their images to be blocked in future versions of the AI model.

DALL-E 3 will be initially available to ChatGPT Plus and ChatGPT Enterprise users in October, with a gradual rollout to research labs and the API service later in the fall. A free public version release date has not been confirmed.