OpenAI has expanded ChatGPT’s capabilities by integrating image generation powered by the GPT-4o model. Now, users can create high-quality images simply by entering text descriptions in the chat.
Previously, this feature was only available to paid subscribers, but it is now free for all users, with a limit of three images per day.
The GPT-4o model, which powers this feature, can accurately depict text and follow complex instructions using its knowledge base and chat context. This allows for highly detailed and precise image generation.
The new feature has sparked significant interest among users. Many have started creating animated portraits in the style of Studio Ghibli, leading to a surge in social media activity.
To ensure transparency and protect copyright, OpenAI has implemented C2PA metadata in all generated images. This helps distinguish AI-generated images from human-created ones and prevents potential copyright violations.
However, the feature has also faced some controversy. Users discovered that ChatGPT refuses to generate images of "sexy women" but does create "sexy men." OpenAI CEO Sam Altman acknowledged this issue and promised a fix.
Overall, the expansion of ChatGPT’s image generation capabilities with GPT-4o marks a major step forward in AI development, making it more accessible and functional for a wide range of users.