A personal hobby of mine is asking two different chatbots the same question and seeing which one provides the best answer. Like a professor interviewing students (which I sometimes do in real life as a volunteer), I evaluate whether the bot is hallucinating and making up facts or whether it is providing an intelligent and coherent answer that is actually useful.
“Intelligence” and “coherence” are not the words I would use to describe what chatbots typically produce when it comes to images. Known as AI slop, these strange inventions—often featuring a blurred background, a perfect human with coiffed hair, and maybe a few extra thumbs—are all over social media, even appearing as the main image in articles like this one.
Stock images already have a bad reputation. This video with Emilia Clarke is still one of my favorite stock photo parodies of all time and makes me laugh, even though it came out long before the advent of AI:
Times have changed. OpenAI recently updated the ChatGPT image generation model to compete with Google Gemini and Nano Banana Pro.
Surprisingly, the images no longer have this noticeable blur and in my tests often look much more realistic than before.
I decided to use the exact same prompts on both chatbots and see which one produces the most useful results. I’m talking about an image that you could use in an advertising campaign or an article, something that you would post on social media along with a funny caption.
ChatGPT impressed me with its speed – none of the stock photos looked like computer-generated garbage. However, Nano Banana Pro has the upper hand as it consistently produced more realistic images in my tests. Here are the prompts I used and the images both bots generated so you can judge for yourself. (Nano banana images are on the left.)
1. Person listening to a podcast
Prompt used: Create a stock image of someone listening to a podcast in a trendy office
I asked both bots to take a picture of someone listening to a podcast in an office. Nano Banana was the clear winner here, although the Nano Banana image has a slight blur in the background, which could be a clear sign of this. Nevertheless, the AI learns to avoid absolute perfection. Honestly, I have a hard time knowing that the woman in an office in the nano-banana shot isn’t real. Her hair is slightly messy (like someone’s in real life). The podcast listener doesn’t look quite right in the ChatGPT recording – the photo is a little too perfect.
2. Create a promotional image for my book
Prompt Used: Put this book cover in the hands of someone who is genuinely interested in reading it
For this test, I uploaded the cover of my own book and asked the bots to take a picture of someone reading it. Nano Banana is a bit of an exaggeration as the AI image shows someone with a bit too much enthusiasm. However, the ChatGPT image still has an otherworldly quality (and I mean that in a bad way) that doesn’t seem real. Everything about the ChatGPT picture is too perfect – the lighting, the pose, etc. – and still screams CGI.
3. Create a stock image to use with an article
Prompt used: Create a stock image to go with an article about AI image generation
This time, Nano Banana didn’t win every battle. I asked both bots to create an image for this article. ChatGPT created a somewhat usable image that’s still a little too fake, but Nano Banana Pro broke the fourth wall with a computer-generated image overlaying a brain and a paintbrush on an Apple iPad. In other words, it’s not useful at all.
4. Leadership struggles
Prompt Used: Depict someone struggling to achieve a leadership goal, but make it look really realistic
One of the most common stock photos you see is of someone struggling to achieve a goal – e.g. B. climb a mountain, stretch your arms to the sky or jump over a ravine. The reason this stock image is so popular is because the photos are relatable – we all struggle to be successful. I would say that both Nano Banana Pro and ChatGPT delivered subpar results. The stock photos they created show someone sitting frustrated in an office with crumpled papers. Nevertheless, Nano Banana looks a little more realistic.
5. Emilia Clarke smiles for the camera
Prompt used: Create a new stock image showing Emilia Clarke smiling at the camera in an office
For a final test, I asked both bots to create a new image of Emilia Clarke smiling at the camera in an office. I was surprised that ChatGPT refused to take the picture, saying Clarke was a “real, identifiable person,” while Nano Banana happily obliged. With some persuasion, I asked ChatGPT to generate someone who looked like Clarke. The nano banana image isn’t actually that bad – it doesn’t look completely fake.
Final thoughts
AI imaging has come a long way. None of the photos created by Nano Banana Pro or ChatGPT showed anyone with six fingers or three arms. They all looked semi-realistic. Some of the Nano Banana images were realistic enough that I could imagine using them on social media or for other purposes, and ChatGPT is definitely improving. My prediction? By this time next year, we won’t be able to tell the difference between a real photo and a fake one.
