But like many early AI systems, DALL-E 2 struggled with consistent text rendering, often producing garbled words and phrases within images. It also had limitations in following complex prompts with multiple elements,

OpenAI has rolled out a major AI image generation upgrade with the GPT-4o model. Here's what it can (and still can't) do.

The model can depict legible text and is more practical than the surrealistic generators of the past. But who will use it?