Quick links
It’s easy to think that AI can create any type of image, but these tools have their limitations. To test this, I took a series of smartphone photos to prove that AI can go no further, some focusing on aesthetics, others on practicality.
1 Explosion photo series
While walking one morning, I took a series of time-lapse photos on my iPhone of people riding bikes along the bridge. I wanted this to look a bit like a magazine or comic strip and show a specific sequence of events. As you can see, I was able to capture every few seconds between bikes crossing along the bridge.
Although AI is good at creating individual images, it is not really capable of creating a series of images. For example, ChatGPT often changes the entire image when I ask it to make minor edits. Furthermore, using the same prompt often yields completely different results.
2 Photography marks
I’ve talked in depth about generative AI being bad at generating text in images. Although I sometimes get good results when using programs like ChatGPT, these tools are often way off the mark. I’ve had numerous experiences where text was unreadable, so I won’t be relying on generative AI for this type of image any time soon.
Trying to create text in AI images made me realize how I often take the simplicity and usefulness of smartphone cameras for granted. When I want to take a photo of a sign, all I have to do is point my camera at it and take the photo.
However, I also wanted to make the image more interesting. So, I made the tag part of the introduction. Generative AI tools can be good at creating bokeh and background effects, but since they can’t generate text, these types of photos are better when taken on a phone.
3 Leading lines
The generative AI is very good at creating landscape-style images, but I’ve sometimes found that it’s lacking when I’m trying to create images with more complex indents. It’s fine for straight roads, but beyond that, the AI tends to struggle. On the other hand, it is much easier to achieve leading lines using your smartphone.
In this example, I took a photo at the beach while standing on the boardwalk. As you can see, I simultaneously took a highlight shot with my smartphone. I liked the way the light hits the walkways, and it’s very cool how the walkway starts to curve to the left as it reaches the background.
I could probably create a similar image using generative AI, but it wouldn’t do the basic mainline element justice. However, you may want to try out some basic AI tips and tricks to improve your chances of getting the result you want.
4 Night photography with smartphones
The generative AI can take some interesting night shots, but it tends to fail to get the colors right. In my opinion, generative AI is better for daytime images. In this nighttime smartphone photo, I specifically wanted to capture an interesting-looking building. This type of architecture is typically more difficult for a generative AI tool to replicate.
The light reflections in this image are also unique. While AI can create reflections in water, it can make them look more like watercolor paintings than the real reflections you see in photos. To take better night photos on your smartphone, consider trying these night photography apps.
5 Silhouettes
Capturing silhouettes can be quite difficult, and to be honest, it often happens to me by luck rather than actually trying. Silhouette photography isn’t just about making your subject appear darker, it also needs light to hit your image properly. For example, in the image below, the light hits the modern building on the left at just the right angle.
It’s also difficult for AI to simulate different shades in the sky for silhouettes, even with specific prompts. Furthermore, I feel that these tools often overestimate shadows; If I wanted my photos to look like this, I could just use the photo editing app on my phone.
6 Macro photography
AI can be good at taking close-up shots, but with some limitations. Although these tools are great for creating images of people, I find they struggle with images of plants. To test this, I took several macro photos of flowers and leaves.
My smartphone camera was able to capture the different textures of these plants, which is exactly what I wanted. It was also good at capturing true colors without over-saturating them; Generative AI tools tend to do this using bright colors. However, as I discovered when trying macro photography with my smartphone, this technique takes a lot of practice.
7 Crouching down
I’m always trying to find new angles in my smartphone photography, and bending down is one way to do that. I can ask the AI platform to create an image that looks like I did that, but it barely gets it right. Since I like to have more control over my composition, doing it with my smartphone is easier.
The marks in this photo were my main area of focus, as I first went under the bridge before crouching down further. As a result, I was also able to get enough of the sky in my photo. Asking an AI tool to do this probably wouldn’t give me what I wanted.
8 Frames within frames
The AI is good at creating images that only have one or two layers, but things get more complicated when I try to add more elements to the image. For example, I like to capture frames within frames; ChatGPT holds its own in this field, but the results are far from perfect. To try to do something that AI can’t do, I took the image below.
I wanted to include the car in the distance, something the AI would have encountered. Since it often has difficulty dealing with different patterns, I don’t think it would have been successful at creating lines on either side of the bridge either – the AI tends to be better at curves.
AI has its advantages, but it cannot be relied upon for all types of images. Smartphone cameras are often outperformed by AI, and although this may change in the future, I still find smartphone photos superior.
