Many users who want to create image content from text descriptions are waiting to try OpenAI’s Dall-E system. But whether the technology will pass the corporate compliance meeting could be a determining factor in its ultimate success in the company.
The San Francisco-based AI research lab said on July 20 that it will make the tool available to more users on its million-strong waiting list in the coming weeks. Dall-E, which uses natural language to translate text into images, is now available in beta.
The disclosure came a few months after OpenAI released the second version of Dall-E, Dall-E 2. The research lab said Dall-E 2 can create more realistic images than the original Dall-E, which first introduced in 2021 as a 12 billion parameter version of GPT-3.
Business use cases
While many consumers have successfully used Dall-E to create creative memes or even create music videos, businesses can also find success with the tool, said R “Ray” Wang, an analyst at Constellation Research.
“It’s about accelerating, automating and using AI to create content,” Wang said.
The tool is useful for companies working in the commercial printing industry or those considering creating content for the metaverse.
While there are different ways to create content at scale right now, most of them are manual, Wang said. Dall-E automates this manual process.
It’s about accelerating, automating and using AI to create content. Ray WangAnalyst, Constellation Research
In the current iteration of the technology, OpenAI will allow Dall-E users to reprint, sell and market their images.
Responsible AI and Ethics
In this release, OpenAI takes a responsible approach to AI, Wang said.
The research lab said its filters are more effective at preventing users from creating violent, adult or political content. However, OpenAI needs to show the market how it built the models with the ability to audit the dataset and results, said Andy Thurai, another analyst at Constellation Research.
An AI program, written in plain text, can be instructed to maliciously edit an image for posting online, potentially damaging a company’s or an individual’s reputation, Thurai said. He added that users should exercise caution and provide governance when displaying a generated image versus an artist’s original work. Additionally, OpenAI must demonstrate that it can only handle G-rated images.
The ethical and privacy implications of Dall-E may make companies skeptical of using the tool, because the tool may misrepresent what is originally said. “Enterprises are less tolerant of mistakes,” said Gartner analyst Sid Nag.
So while Dall-E could be a real game changer, compliance regulations could prevent this technology from ever making it into the company’s door.
OpenAI against the competition
This trial version of Dall-E puts OpenAI one step ahead of competitors like Google, which launched Imagen earlier this year. Wang said vendors are eager to get these products into the hands of users so they can see how they will use the technology.
“This is really a speed game,” he said. “This is about building the training data on the back end to see how people use it [and] what else can they do to modify or add features.”
Competitors such as Google will likely see how the market receives OpenAI before following suit, Wang said.
Dall-E users receive free monthly credits. Anyone who wants more credits can purchase an additional 115 credits for $15, allowing them to create 460 images.