"The images in ChatGPT were much more popular than we expected (and our expectations were pretty high)," Altman said in a post on X on Wednesday.
On Tuesday, the company announced the launch of GPT-4o's native image generation feature, allowing users to upload and modify images, and said the feature will soon be available for all tiers of ChatGPT. As of Wednesday, OpenAI had rolled out the feature only to ChatGPTPro, Plus and Teams subscribers.
OpenAI has faced similar problems with product launches in recent months, and has repeatedly blamed a lack of computing power. Shortly after launching Sora last December, OpenAI disabled the registration feature. Perhaps the company's planned $500 billion Stargate data center project will help address these capacity constraints.
Related articles: