On Sunday, OpenAI CEO Sam Altman urged users to reduce their usage of ChatGPT’s latest image generation feature, citing overwhelming demand.
In a post on social media platform X, he said, “Can yall please chill on generating images? This is insane. Our team needs sleep (sic).”
Despite his appeal, many users were unconvinced, some even suggesting he should fire his team.
Responding to one such post, Altman defended his team, saying they had built one of the biggest websites in the world in just 2.33 years.
“No thanks. In addition to building AGI, this team is on a trajectory to build the biggest website in the world from a cold start 2.33 years ago,” Altman replied to the post.
On March 25, OpenAI introduced its newest AI-powered image generator, integrated with ChatGPT’s GPT-4o model.
This feature allows users to refine images iteratively and maintain consistency across multiple edits.
Since its launch, there has been a surge in user engagement, with many experimenting by transforming images into various artistic styles, including Studio Ghibli.
However, the rapid adoption has put a strain on OpenAI’s infrastructure. Altman acknowledged the situation: “It’s super fun seeing people love images in ChatGPT, but our GPUs are melting.”
Due to copyright concerns, OpenAI has also restricted the generation of images in the style of certain living artists and studios, such as Studio Ghibli.
To manage the high demand, OpenAI plans to introduce temporary rate limits. The free version of ChatGPT will soon have a cap of three image generations per day.
Altman assured users that these restrictions are temporary and that the company is working to improve efficiency.
Unlike earlier AI-generated art that relied on external tools like OpenAI’s DALL-E, the new image generation feature is built directly into ChatGPT-4o.
Users can create visuals seamlessly using text prompts and specify details like aspect ratio, color schemes, and transparent backgrounds.
The model can also analyze and learn from uploaded images to enhance its output. In a blog post, OpenAI explained that their models have been trained on a vast collection of online images and text, allowing them to understand how images relate to language and how they connect with each other.
