Since launching in November 2022, like many other people, I have been using and testing OpenAI’s ChatGPT. It is an Artificial Intelligence (AI) chatbot capable of providing coherent answers to questions and generating detailed suggestions to almost anything that is asked of it.
I was going to write this blog myself but curious to test ChatGPT’s ability to create content, I put the task to the AI – perhaps not the most sophisticated test but I feel the question was challenging, with minimal information provided.
I asked ChatGPT to compare AI and Formula 1 – two subject matters that at first may not appear to have too much in common – and I was quite impressed with the result.
If you think about it, Formula 1 and software development do share many similarities, AI even more so. Small and frequent changes are applied daily, team effort overall with individuals making big impact, tons of research and development go into the product, the use of data models are a must, lots of quality assurance, and if you don’t keep up with the latest advancements you’re left behind.
Interestingly, F1 and Artificial Intelligence started around the same time in the 1950s – you could say it’s taken 73 years for a F1 car to get around Silverstone in 1:27:36.9 and for software to become advanced enough to be able to produce the blog below off a vague description. Comparatively in 1950, the fastest lap around Silverstone was 2:13:23.6, and in 1956 the Samuel Checker AI was just about able to play a game of checkers on the IBM 701 computer – which was a rather large machine, taking up a whole room!
Back to the experiment. Here’s the post requirements that were asked (with intentional bad grammar and spelling):
Write a short story between 200 and 250 words on how AI and Formula 1 are similar because both are high performance need improvement where they Start at the year isn’t where they finish at the end if you don’t improve and update you fale.
And here’s ChatGPT’s unedited output, which it wrote in less than 30 seconds:
As the sun rose over the track, the team of engineers huddled around the sleek, silver race car. They had been working tirelessly for months, pouring over data and fine-tuning every last detail to ensure that their vehicle was the best it could be.
As they prepared to take to the track, they knew that they were up against fierce competition. The world of Formula 1 was filled with teams who were all striving for the same goal – to be the fastest and the best.
But the team was confident in their car and their abilities. They had spent countless hours analyzing data, making adjustments, and testing new technologies. They knew that if they were going to succeed, they needed to be constantly improving and updating their car to stay ahead of the competition.
As the race began, the team watched nervously from the pits as their driver hurtled around the track at breakneck speeds. They knew that every second counted, and that even the slightest mistake could mean the difference between victory and defeat.
In the end, their hard work paid off. As the checkered flag waved, their car crossed the finish line in first place, triumphant and triumphant.
In many ways, the world of artificial intelligence is not so different from the world of Formula 1. Both are fields that are constantly evolving and improving, and both require teams of experts to work tirelessly to stay ahead of the competition. And just like in Formula 1, those who fail to keep up with the latest advancements will inevitably fall behind. The end.
When I started reading this, I was really wondering where the ChatGPT AI was going and if it understood the requirements correctly, given the heavy narrative, scene setting nature of the writing. There is a glimmer of the intended assignment in the final paragraph, with ChatGPT offering a loose comparison between F1 and AI. The piece is signed off with “The end”, typical of closing a story.
Does this make it a bad piece of writing? Taken in the context that it is AI generated content, not at all. The AI manages to paint a picture for the reader, telling the story of a hopeful F1 team who have laboured over their race car in pursuit of victory – it’s a complete short story, but I think the AI leaned too far into that, missing the point of comparing the two subjects.
So, what could be changed? I didn’t want to be too much more specific, as this is meant to challenge the ChatGPT AI. I asked the exact same question as above, but this time requested a ‘blog post’, rather than a ‘short story’.
Here’s the unedited result:
Artificial intelligence (AI) and Formula 1 racing have a lot in common in that they both rely on high performance and the constant need for improvement. Just like a Formula 1 team, an AI system is only as good as its current capabilities. If an AI system is not constantly learning and improving, it will fall behind and become obsolete.
In Formula 1, teams are constantly analyzing data, testing new technologies, and making adjustments to their cars in order to gain a competitive edge. Similarly, AI systems also rely on data and the ability to learn and adapt in order to improve.
Both AI and Formula 1 are also constantly evolving industries. New technologies and techniques are constantly being developed and implemented, and those that do not keep up with the times risk falling behind. In both cases, success is dependent on the ability to continuously learn, adapt, and improve.
One key difference between AI and Formula 1 is that while a Formula 1 team may have a set goal or objective, such as winning a race or championship, the goals of an AI system can be more open-ended. AI systems can be used to solve a wide range of problems and can be trained to perform a variety of tasks.
Overall, AI and Formula 1 are both high performance industries that require continuous improvement in order to stay at the top of their game. Whether it’s analyzing data, testing new technologies, or adapting to changing conditions, both AI and Formula 1 rely on the ability to learn and evolve in order to succeed.
This time, with a slight change in requirement, ChatGPT has provided an informative piece of content giving a clear introduction, several direct comparisons that offer both reasoned similarities and differences, and a conclusion that references the points made prior. While this response is to the point, more academic than creative, and doesn’t conjure up visual imagery; it provides the reader with concise information and key take aways.
Does this make it the better piece of content? In my opinion, yes and no. While it’s certainly more informative and gives a much better argument for the comparison of F1 and AI, it does lack the charm of the ‘short story’ response.
What has been provided by both responses is solid starting points for an article to be built off. By combining the ‘short story’ and ‘blog post’ responses, a compelling article that draws the reader in with the vision of a Formula 1 team preparing to race and a software development team tinkering with the next big leap in AI could be written, interwoven with informational pieces to show how these two disparate groups may not be so different.
This blog writing experiment has shown me just how far AI has come in offering a solid alternative to researching a topic, for providing ideas and ultimately writing content, and that the first response may not be the definitive answer with follow up questions continuing to have their place.
While ChatGPT won’t be replacing content creators, it certainly shows potential for making creatives more efficient and even help combat writer’s block by providing starting points for blogs, social media posts, stories, and beyond.
This decade looks promising for AI… let’s see what it brings, changes, and disrupts
As a further extension of this experiment, head over to the MIC Global Blog to read an article written by combining the two responses given above.