ChatGPT currently allows users to upload images for analysis and editing, but there appears to be no documented way for the assistant to redisplay those original uploaded images inline within its own responses.
This would be particularly useful for visual creative workflows involving character design, illustration, image comparison, and iterative editing.
For example, users should be able to ask ChatGPT to:
-
Redisplay a previously uploaded reference image.
-
Compare multiple uploaded images side by side.
-
Display the original image alongside an edited or newly generated version.
-
Reference specific images visually while explaining design differences.
Ideally, ChatGPT should be able to reuse the existing attachment reference without regenerating, modifying, or externally hosting the original image.
This functionality would significantly improve image-based conversations and creative collaboration, particularly in the ChatGPT web interface.