DALL·E 2, OpenAI’s artificial intelligence-powered image generator that creates realistic images and art from a description in natural language, is now available as a beta to the public in a limited capacity. Until now, the powerful AI tool was available only for a select group of users after it was announced in April.
The San Francisco–based company on Wednesday announced that it will soon be sending out invites to one million people from its waitlist over the coming weeks for its latest version.
For the unversed, DALL·E 2 takes a phrase or string of words as input and returns professional-looking paintings or hyper-realistic photographs as its output. The DALL·E 2 also includes the below features:
Edit: This allows users to make realistic and context-aware edits to images they generate with DALL·E or images they upload using a natural language description.
Variations: This can take an image generated by DALL·E or an image uploaded by a user and create different variations of it inspired by the original.
My Collection: This allows users to save generations right in the DALL·E platform.
“DALL·E allows users to create quickly and easily, and artists and creative professionals are using DALL·E to inspire and accelerate their creative processes. We’ve already seen people use DALL·E to make music videos for young cancer patients, create magazine covers, and bring novel concepts to life,” the company writes in a blog post.
Every DALL·E user will receive 50 free credits for their first month of use and additional 15 free credits every subsequent month. Each credit represents four pictures based on one original DALL·E prompt, or three images if the user offers an edit or a variation prompt.
Users can also buy additional DALL·E credits in 115-credit increments (460 images) for $15 on top of their free monthly credits. Each time a prompt is entered and a user hits “generate” or “variations,” one credit is applied.
Further, artists who are in need of financial assistance will also be able to apply for subsidized access, OpenAI says.
The beta version also allows users to get full usage rights to commercialize the images they create with DALL·E, including the right to reprint, sell, and merchandise. This includes images they generated during the research preview.
OpenAI says that it has already received interest from users who are planning to use DALL·E images for commercial projects, like illustrations for children’s books, art for newsletters, concept art and characters for games, moodboards for design consulting, and storyboards for movies.
OpenAI adds that it has worked with researchers, artists, developers, and other users to ensure that its technology is not misused for harmful purposes or to create misinterpretation and deep fakes.
The company says its taking steps to reduce the creation of deceptive content by rejecting image uploads containing realistic faces. It has also made its content filters more accurate which does not allow users to generate violent, adult, or political content, among other categories. Additionally, it has implemented a new technique in DALL·E which creates images of people that more accurately reflect the diversity of the world’s population and does not specify race or gender.
OpenAI emphasizes that the DALL-E 2 will continue to have automated and human monitoring systems to safeguard against misuse.
“Expanding access is an important part of our deploying AI systems responsibly because it allows us to learn more about real-world use and continue to iterate on our safety systems,” OpenAI added. “We are continuing to research how AI systems, like DALL-E, might reflect biases in its training data and different ways we can address them.”