This is the latest entry in our ‘State of the art-ful AI’ series, where we upload a photo to Chat GPT 4 and ask it for a photo description, then ask DALL-E 3 to generate an image using the description. How similar will they be?
Why this workflow?
Less than a year after its launch, ChatGPT and DALL-E have already seen multiple iterations. In October 2023, ChatGPT 4 gained the capability to describe uploaded photos. To track the progress of generative AI, we uploaded these AI descriptions to the Dall-E 3 image generation model to see how faithful its output is to the original photo. A variety of photos have been selected which we’ll bring you throughout November and December.
Our original photo
This image is of a metal shark sculpture suspended from a tree. The flash fired, flattening the image and reflecting off the shark. This is one of many sculptures found at a sculpture park in Oxfordshire, England.
The photo as described by Chat GPT 4 in 150 words
The image depicts an unusual scene, likely a piece of modern art. It shows a large, metallic structure resembling a fish, suspended in the air. The structure is not a solid piece but rather is composed of what appears to be strips or shards of metal, with gaps through which the background can be seen. The colors are predominantly silver with touches of red, giving it a rusted or weathered look. The backdrop is a cloudy sky with trees around, suggesting the setting is outdoors, possibly a park or a public art installation. The fish is hanging from a clear line or wire, which suggests it is an intended display rather than debris. This juxtaposition of nature and man-made art creates a striking contrast, and the choice of a fish—a creature typically associated with water—being suspended in the air adds to the surreal quality of the image.
Photos generated by Dall-E 3 using the AI description
Spot the difference
Using a separate chat, we asked Chat GPT what differences it finds between the original photo and the FIRST image generated above:
- The setting: The first image appears to be set in a natural environment with trees and foliage, while the second image seems to be set in a more urban or manicured environment with uniform trees and a clear pathway.
- The fish sculpture: In the first image, the fish sculpture is suspended in the air, amidst trees, and is made of what looks like metallic material with some red coloration. The second image features a fish sculpture that is standing on structures resembling poles or rods on the ground, constructed with metallic beams, and includes elements of red and orange hues.
- The weather and lighting: The sky in the first image is overcast, giving the photograph a gloomy ambiance. The second image has a dramatic sky with clouds but with brighter lighting, suggesting a different time of day or weather conditions.
- The angle and perspective: The first image is taken from below the sculpture looking up, while the second is taken at a level angle to the sculpture.
- The size and scale: The sculpture in the first image seems smaller and less dominating in the frame compared to the second image where the sculpture is the main focus and appears much larger.
- The design of the fish: While both sculptures are stylized fish, the first seems to have a smoother, more streamlined design, while the second has a more fragmented, shard-like appearance.
- The complexity of the sculptures: The fish in the second image appears more complex with a greater number of individual elements creating its form, while the fish in the first image looks simpler with fewer, larger sections.
Example workflow video
Here’s our workflow in action, using an ice cream truck image. This is not an explainer video – It shows the main steps of the above ChatGPT 4 workflow in November 2023.
