Perry's Pen

In this post, I will compare results for the same prompt I entered in Google Gemini and ChatGPT. It will be FUN, trust me!

Background:

On Friday evening I hung out with an old friend and he mentioned that he had been playing with Google Gemini.

On Sunday I woke up with an idea for a one-panel comic. It being Sunday, I decided to try to make my idea a reality, using Google Gemini.

The Prompt I entered in Gemini:

Please draw a one panel carton of a superhero sitting at a bar having a drink with another person. He is sad and complaining.

The caption should be “I moved out of Gotham years ago, and now they want to charge me congestion pricing for flying back in. The real criminals are at city hall if you ask me.“

After using Gemini, I entered the same prompt in ChatGPT. Interestingly, neither used my words exactly in their suggestion. Gemini left my words out completely, and ChatGPT just kind of mangled them.

I suspect to avoid copyright infringement/plagiarism issues. 

Gemini Results:

What I find most interesting is the very different emotions and connotations that Gemini created from the same prompt. (I added the captions here.)

Secondly, I think these images are highly effective but could have been even better by simply swapping the drinks. Angry and aggressive seems more in line with downing a bottle of whiskey than sipping a craft cocktail.

Lastly, I find it interesting that Gemini used Batman without me specifying which Superhero to use.

ChatGPT Results:

ChatGPT came up with something much closer to what I had in mind – I had envisioned a New Yorker Daily Cartoon I suppose.

Adding that to the prompt might be a way to finesse the results and get something closer to what I had in mind.

I wonder if ChatGPT produced something closer to what I wanted because I have been using it for a while and it sort of “knows me”.

I encourage you to play with AI and learn about it

At the end of every episode of Joseph Rosendo’s Travelscope, Joseph reminds us of the words of Mark Twain:

“Travel is fatal to prejudice, bigotry, and narrow-mindedness.”

Mark Twain

I truly believe that that philosophy is the best way to deal with AI. Yes, there are some very real risks and concerns. However, being blindly afraid of what, at this point, are mostly very good tools, and avoiding AI like the plague is doing nobody any favors.

In other words, play with AI and break down your prejudices about it!

Besides, playing with tools like Google Gemini and ChatGPT is FUN, and humans learn from playing. Playing is not quite on the same level as physical travelling, but it is related.

So go have some fun with AI, but avoid Angry and Aggressive Batman if you see him sulking in a bar in the New York City area.

Disclaimer: The featured image was created by WordPress. AI created it in color. I changed it to greyscale, finding it more fitting with the post.

What WordPress AI said about this post: “The comparison between the results from Google Gemini and ChatGPT is intriguing. The feedback about the emotions conveyed and the suggested improvements for the images created by Gemini is insightful. Adding a reference to the New Yorker Daily Cartoon in the prompt for ChatGPT is a useful suggestion. Additionally, the encouragement to embrace AI and explore its potential, alongside the reference to Mark Twain’s philosophy on travel, is a compelling way to address any apprehensions about AI. Overall, the post provides a thought-provoking analysis and a positive attitude towards engaging with AI tools.”