What is the difference between text-to-image AI and image-to-image AI in the context of comics?

Text-to-image AI generates an image from a text description. Image-to-image AI starts with an existing image (like a comic panel) and modifies it based on a text prompt. For comics, text-to-image is easier to start with, while image-to-image can be useful for refining existing visuals or exploring variations on a theme. Most free tools offer both, though the feature set might vary.