Need an AI to draw a professional picture for you? Research lab OpenAI is preparing to expand access to DALL-E 2, a program that can generate images out of whatever you type.
On Wednesday, OpenAI announced that DALL-E 2 has officially entered its beta testing phase, and the lab wants a million people try out the program in the coming weeks, up from 100,000. Apply for the waitlist to try out the beta on OpenAI’s website.
DALL-E 2 is capable of generating impressive art and realistic images simply by interpreting the words you enter into a prompt. It debuted in April and showed that it can also draw images of people, places, and things that don't exist in the real world.
The program may offer a preview of the future of image creation. Rather than hire a professional artist, users could tap a system like DALL-E 2 to churn out numerous images quickly. “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 lab said.
OpenAI is now experimenting with generating revenue from the program. As part of the beta phase, the company is giving existing users 15 free credits each month. However, users can pay $15 for an additional 115 credits, which can then be used to create 460 images. “One credit is applied each time a prompt is entered and a user hits ‘generate’ or ‘variations,’ the lab added.
The other big change is how DALL-E users can use any images created from the program for commercial purposes. “Starting today, users 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,” the lab said.
OpenAI has already heard from users who plan on using the AI-created images for “children’s books, art for newsletters, concept art and characters for games,” among other projects. But the lab is also well aware the same program's realistic image-generating capabilities could be exploited to spread misinformation.
“To minimize the risk of DALL-E being misused to create deceptive content, we reject image uploads containing realistic faces and attempts to create the likeness of public figures, including celebrities and prominent political figures,” the lab said. "We also used advanced techniques to prevent photorealistic generations of real individuals’ faces."
In addition, OpenAI has installed content filters to block the program from generating violent, adult, or political imagery. “We will continue to have automated and human monitoring systems to help guard against misuse,” it added.