OpenAI, now under Microsoft's umbrella, has introduced an upgraded version of its text-to-image tool known as 'DALL·E 3,' powered by the AI chatbot ChatGPT. Currently in research preview, DALL·E 3 will be accessible to ChatGPT Plus and Enterprise subscribers in October through the API and in Labs later this autumn.
DALL·E 3 is uniquely integrated with ChatGPT, enabling users to leverage ChatGPT as both a brainstorming partner and a prompt refiner. By simply expressing their desired image concept, users can receive customized, detailed prompts from ChatGPT tailored to DALL·E 3. Should users want to make slight modifications to an image, they can instruct ChatGPT with just a few words.
OpenAI claims that DALL·E 3 exhibits an enhanced ability to understand nuance and detail, making it proficient in transforming users' ideas into highly accurate images. In addition, OpenAI assures users that the images they create with DALL·E 3 remain their property, granting them the freedom to reprint, sell, or merchandise these images without seeking OpenAI's permission.
Furthermore, OpenAI has implemented safeguards to restrict DALL·E 3's capacity to generate violent, adult, or hateful content, a concern that had arisen with previous versions of the tool. DALL·E 3 has also been programmed to decline requests for images mimicking the style of living artists. Creators can now opt to exclude their images from future image generation model training by OpenAI.
This integration of DALL·E 3 with ChatGPT offers users a seamless and creative way to generate highly detailed and tailored images from textual prompts, advancing the potential for AI-generated content in various applications.
Inputs from IANS