Is OpenAI’s new image generator a game-changer for advertising creatives? It depends who you ask.
On Tuesday, OpenAI released what it called its “most advanced image generator yet,” built into the latest version of ChatGPT. Called “4o Image Generation,” the tool sparked so much buzz and attention, activity on it overloaded the company’s servers, according to OpenAI CEO Sam Altman.
With 4o Image Generation, a user can enter prompts into ChatGPT and produce images like art, diagrams, infographics, logos, and stock photos. This week, some users experimented with the generator to create images in the animation styles of South Park, The Simpsons, Rick and Morty, and Studio Ghibli—the latter of which became a viral trend on social media.
OpenAI’s previous text-to-image generation tool, DALL-E, is a separate system from ChatGPT and users have reported pitfalls in the quality of its images. But 4o Image Generation, built into a single system with ChatGPT, has already received praise for being more sophisticated and creating more realistic and accurate imagery.
With all the hype, what does 4o Image Generator mean for creatives in advertising? ADWEEK asked a few who use AI tools for their views.
‘Gives power to the masses’
Wesley ter Haar, co-founder and chief AI officer at Monks, said the new image generator is “amazing” because of its accessibility.
“It gives a level of creative quality and control to the masses. Previously, this was only available to power users with more convoluted set-ups and workflows,” he said.
While he cautioned advertising creatives about embracing AI tools without considering brand safety concerns for their clients, overall “this development clearly shows where the technology—and because of that our industry—will be headed.”
While Andrés Ordóñez, global CCO of FCB, agreed that the 4o Image Generator is “a big step forward” in terms of “how seamlessly it integrates into the creative process,” the AI program Midjourney “still leads the way” in terms of visual craft, he said.
“[Midjourney’s] Editor tools let you zoom into a specific part of an image, refine it with precision, and maintain visual consistency. It also handles stylized, expressive imagery with a trained eye,” Ordóñez explained.