Uncategorized

Behind the Scenes: How we Wrote a Chamber of Commerce Article using ChatGPT

June 30, 2023

SpinDance was honored to contribute an article to the July, 2023 issue of The West Coast Way, the magazine for The Michigan West Coast Chamber of Commerce. 

The article, “Demystifying Machine Learning: How AI is Shaping the Business Landscape” had a not-so-secret twist: The entire article, apart from the postscript, was written by ChatGPT. 

This blog post explains the backstory of how we “wrote” the article. We also threw in some fun variations of the article at the end. 

Let’s dive in.

The “Writing” Process

We used ChatGPT 4.0 to compose the article. ChatGPT is a family of machine learning models that generate text given a human-written prompt. The 4.0 version of ChatGPT is one of the most advanced technologies in this space. 

The advanced math that makes these models work is outside the scope of this blog post. However, if you’re curious, I suggest you read the article “What Is ChatGPT Doing.. And Why Does It Work?” By Stephen Wolfram. It does a great job explaining the process in great detail, without using too much math.

My first step in writing the article was to ask ChatGPT this question:

“What are 5 good ways companies can start exploring artificial intelligence. Consider ideas that are good for both small, medium and large businesses. “

It returned five examples, four of which we used in the final essay. 

I then appended these examples to a second question:

Write an 600 word article on machine learning and the application of it for small business. The target audience are business leaders who are unfamiliar with machine learning. The article should include:

  1. An introduction.
  2. A brief history of machine learning.
  3. An explanation of recent advancements in generative AI
  4. 3 to 5 takeaways in how businesses can leverage AI.
  5. A conclusion.

Voice and style guide: Write in a conversational, relatable style as if you were explaining something to a colleague. 

It produced the first draft of the article in about 30 seconds. It was a good, but not perfect, first draft. I copy-and-pasted the article in a working document.

Next, I asked the chatbot for 4 titles. It responded with these:

  1. “Unlocking Business Success with Machine Learning: A Comprehensive Guide”
  2. “Machine Learning for Business Leaders: Transforming Your Enterprise with AI”
  3. “Demystifying Machine Learning: How AI is Shaping the Business Landscape”
  4. “Harnessing the Power of Machine Learning: A Roadmap for Business Innovators”

We ultimately decided to use number 3. 

I then spent 5 minutes asking the Chatbot to rewrite a few key paragraphs, including the introduction and conclusion. I did this by asking for 3 variations for each section. I copy-and-pasted the ones I liked best into my working document.

The article I had at that point was a bit too long, so I removed the 5th use case from the original list. I added the postscript, and arrived at the final article:

——————–

Introduction

Imagine having a colleague who could predict your next big business move, always knew just what to say to customers, and never missed a beat in spotting a trend. Sound too good to be true? Not in the realm of machine learning. It’s a corner of artificial intelligence that’s making a real splash in the world of business.

A Brief History of Machine Learning

So, how did machine learning get started? Back in the 1950s, some smart people thought that machines could learn to identify patterns. As time went on, they developed more advanced techniques like decision trees and something called backpropagation. But the big leap happened in the 21st century when we started having lots and lots of data and much better computers. That’s when machine learning really started to change the game.

Recent Advancements in Generative AI

Let’s dive into an area of machine learning that’s really pushing boundaries: ‘Generative AI’. This is a type of machine learning that can create new data – like writing text that sounds incredibly human, creating artwork, or even composing music. It’s like having an AI Picasso or Shakespeare on your team! Recent advancements in this field, such as OpenAI’s GPT-3 and GPT-4, can understand context, write full essays, answer questions, and yes, even tell jokes.

How Businesses Can Harness Machine Learning

Alright, now that we’ve got a bit of background down, let’s dive into the real meat and potatoes – how can your business actually use machine learning? Believe it or not, it’s not as complex or sci-fi as you might think. In fact, there are quite a few ways that businesses, just like yours, are already harnessing the power of machine learning to make things run smoother, smarter, and more efficiently. Let’s break down a few of these:

  1. AI Chatbots: AI-powered chatbots can handle customer service queries, book appointments, and assist with sales. They’re an affordable and efficient solution for small businesses, enabling human employees to focus on more complex tasks.
  2. AI-Powered Analytics: Tools such as Tableau and Power BI use AI to help businesses understand their data. They can generate forecasts, identify trends, and highlight outliers, empowering businesses to make informed, data-driven decisions.
  3. Robotic Process Automation (RPA): RPA tools use AI to automate mundane tasks, ranging from extracting data from emails to automating entire business processes. This can free up valuable time and resources for more strategic tasks.
  4. AI in Marketing and HR: AI can personalize marketing messages, predict customer behavior, and optimize ad spend. Similarly, in HR, AI can screen resumes or schedule interviews, making the hiring process more efficient. 
Conclusion

And there we have it – a whistle-stop tour of the world of machine learning. It’s not just tech talk, it’s a real game-changer that’s redefining how we do business. From AI chatbots to AI-powered analytics, it’s all about working smarter, not harder. And the best part? You don’t need to be a tech wizard to get on board. As we push ahead into a future where AI and humans work together, businesses that embrace these technologies are setting themselves up for success. So, let’s grab the bull by the horns and make the most of what machine learning has to offer.

——————–

The entire process took around 20 minutes. I should note, however, that I had been spending a lot of time over the previous 5 months learning how to use the technology. I’m comfortable navigating the chatbot, and I leaned on a few tips and tricks that I outline below.

(Not) Fooling My Wife.

Before I get to the tips and tricks, there’s one other part of the story I want to share. I’ve been writing professionally for close to 20 years. I’m not a terrible writer, but I doubt I’ll write a bestseller. My wife Amy, on the other hand, is a professional, published author. She is a ghost writer, editor, and professional proofreader. She often proofreads my work, giving me invaluable advice and making me sound much more polished than I actually am.

So, she was not surprised when I asked her to proofread the above article. I handed her my laptop, and, as she often does, she went into editor mode: her eyes narrowed as she began to inspect the work.

About 30 seconds in, she smiled a little smile, and handed the laptop back to me.

“You didn’t write this, did you, The AI wrote it.” she said.

“How did you know?” I asked. I wasn’t surprised she figured it out, but I was surprised how quickly she had done it.  

“Because,” she retorted, “it didn’t have any spelling mistakes, and all the commas were in the right spots.” 

Ouch! ChatGPT is already a better writer than I am. 

Tips & Tricks

While ChatGPT might be a better writer, as you can see from the process above, it needs a good editor. This is why I don’t see generative AI outright replacing jobs in the foreseeable future. A tweet I saw summed it up well: “AI won’t take your job… but someone using AI will.”

I believe that over the next few years, learning how to pilot these chatbots will be increasingly important. They will become invaluable virtual assistants to everyone, but especially knowledge workers. Therefore, learning to use the tools is critical.

To that end, I wanted to share a few tips and tricks I’ve learned in my brief time using chatbots:

  1. Use it as a brainstorming tool. As you can see above, my first step was to ask the chatbot to generate ideas. In this case, I was happy with the 5 ideas it came up with. But often, the ideas need tweaking. I’ll typically start by generating a small number of ideas. Once I’m confident the chatbot is heading in the right direction, I’ll ask it for many more, sometimes to 50 to 100 ideas.
  2. Use it to summarize ideas: Once I’ve “fanned out” with a lot of ideas, I’ll ask it to summarize the content. “Find the higher order themes” is a good prompt to use. I do these to try to distill the essence of the material. These summaries are often better than the raw, original lists.
  3. Ask for variations: ChatGPT can’t read your mind, just like a human writer can’t. You likely won’t like the first, or second variation, of what it produces. But, unlike a human, it is happy to produce many variations. Don’t be afraid to ask for 3, 4, or 5 variations of something. You can then cherry pick the best ones, editing as you go.
  4. Verify facts. The article we wrote doesn’t include a lot of hard facts. But if your articles do rely on facts, make sure you verify them. Large Language Models, like ChatGPT, are statistical models, not real consciousness. Therefore, they can easily “hallucinate” facts. It’s always good to double check the data with some Google searches. 

Thoughts on Attribution in the Age of Chatbots

Technologies like ChatGPT will have a massive impact on our society. This is true of every technological innovation, be it the smartphone, internet, or digital transistor. 

Of the many things we’ll need to wrestle with in the age of AI is that of attribution. 

In the case of this article, we were fully transparent with the Michigan West Coast Chamber. The entire idea was to encourage a conversation about the use and role of AI in our local economy. 

But I’ll be honest: I felt a bit weird putting my name on the article. I definitely acted as the author and editor for the piece. As the author, I originated the topic, and selected the relevant themes. As an editor, I selected the best pieces of prose. 

That said: I don’t feel like I wrote the article. 

This isn’t a new phenomenon. Many books and articles we read today are actually written by someone other than the so-called author. Most, if not all the quotes, attributed to CEOs in press releases are actually written by PR professionals. For years, AI has been used to write up summaries of baseball games. 

But as the amount and quality of AI-generated content increases, I wonder how we’ll deal with this as society. 

For myself, I plan on disclosing when I use chatbots. At least, until the norms become more self evident.  

Alternative Variations

Finally, on a less philosophical point, I wanted to demonstrate how ChatGPT is excellent at mimicking different writing styles. I thought it would be a fun way to end this blog post.

To generate these variations, I took the original article, and asked ChatGPT to rewrite it in a different voice. I went with: William Shakespeare, Yoda, a 1980’s power ballad, and Jive.

Enjoy!

Oh, and if you’re interested in learning more about chatbots or artificial intelligence, drop us a note at hello@spindance.com

——————–

William Shakespeare 

Prologue

Imagine, if thou wilt, a confidante who might foretell thy next grand venture, who always hath the precise words to allay thy patrons, and who never overlooks a budding trend. Doth it sound too fantastical to be true? Not in the realm of machine learning, a facet of the grander artificial intelligence that maketh waves in the bustling marketplace of commerce.

A Brief Chronicle of Machine Learning

But whence came this notion of machine learning? ‘Twas the 1950s when wise minds pondered on the potential of machines to discern patterns. As the sands of time slipped through the hourglass, more sophisticated techniques such as decision trees and the intriguing notion of backpropagation were birthed. Yet, the true quantum leap occurred in the 21st century, when data in copious amounts and computers of superior capability graced us. Thus, machine learning became a formidable player in the game of progress.

Current Progress in Generative AI

Let us delve into a domain of machine learning that truly pusheth boundaries: the ‘Generative AI’. This form of machine learning hath the power to engender new data – be it writing text of humanlike eloquence, crafting works of art, or even composing symphonies. ‘Tis akin to having an AI counterpart of Picasso or yours truly, Shakespeare, within thy cohort! Recent strides in this discipline, such as those made by OpenAI’s GPT-3 and GPT-4, can comprehend context, pen comprehensive essays, answer riddles, and yes, even engage in jesting.

How Commerce Can Harness Machine Learning

Now that we’ve painted the backdrop, let us dig into the meat of the matter – how can thy commerce harness machine learning? Fear not, ’tis not as intricate or far-fetched as thou might imagine. There exist a multitude of ways that ventures, akin to thine own, already make use of machine learning to smoothen, to smarten, and to optimize operations. Let us peruse some examples:

  1. AI Minstrels: AI-powered minstrels, or chatbots, can manage customer queries, schedule appointments, and assist with sales. An economical and efficient solution for the humble merchant, it permits human workers to focus on more complex tasks.
  2. AI-Powered Soothsaying: Tools like Tableau and Power BI employ AI to help ventures decipher their data. They can foretell trends, identify patterns, and spotlight anomalies, enabling ventures to make astute decisions based on data.
  3. Robotic Process Automation (RPA): RPA tools employ AI to automate mundane chores, ranging from gleaning data from letters to automating entire business processes. This liberates precious time and resources for tasks of a strategic nature.
  4. AI in Proclamation and HR: AI can personalize messages to the public, predict customer behavior, and optimize expenditure on advertisements. Likewise, in HR, AI can peruse resumes or schedule dialogues, making the hiring process more efficient.
Epilogue

And there we have it – a brisk journey through the realm of machine learning. ‘Tis not merely technobabble, ’tis a real game-changer that’s reshaping how we conduct commerce. From AI minstrels to AI-powered soothsaying, ’tis all about working more wisely, not harder. The icing on the cake? Thou needst not be a wizard of technology to board this ship. As we venture forth into a future where AI and humans collaborate, ventures that embrace these technologies are laying the groundwork for success. So, let’s seize the day and make the most of what machine learning has to bestow.

——————–

Yoda (From Star Wars)

Introduction, hmm

Visualize, you must. A comrade foreseeing your next great trade venture, always possessing the right words for clients, and missing no trend, they do. Too good to be real, this may sound, hmm? Not in the realm of machine learning, it is not. An aspect of artificial intelligence it is, creating substantial ripples in the business world, it is.

Brief History of Machine Learning, A

So, where did machine learning begin, you ask? In the 1950s, it did, when wise beings believed machines could learn to recognize patterns, they did. As time passed, techniques more advanced, such as decision trees and backpropagation, they developed. But in the 21st century, the giant leap took place, when data in large quantities we began to have and better computers we got. That’s when game-changer, machine learning became.

Recent Advancements in Generative AI, hmm

Explore, we will, an area of machine learning pushing the limits: ‘Generative AI’. A type of machine learning this is, that can create new data – penning text that sounds incredibly human, creating artwork, or even composing music. An AI Picasso or Shakespeare in your team, it is like having! Recent advancements in this field, such as OpenAI’s GPT-3 and GPT-4, context they can understand, full essays they can write, questions they can answer, and yes, even jokes they can tell.

Businesses and Machine Learning, How They Can Intertwine

Now, some background we have covered, to the heart of the matter, we proceed – how can machine learning be used by your business, hmm? Complex or futuristic as you might perceive, it is not. In fact, many ways there are that businesses, like yours, already are using machine learning’s power to make things smoother, smarter, and more efficient. Break down a few, we shall:

  1. AI Chatbots: Manage customer service queries, book appointments, and assist with sales, AI-powered chatbots can. An affordable and efficient solution for small businesses they are, enabling human employees to focus on tasks more complex.
  2. AI-Powered Analytics: Tools such as Tableau and Power BI, AI they use to aid businesses understand their data. Forecasts they can generate, trends they can identify, and outliers they can highlight, empowering businesses to make decisions, data-driven they are.
  3. Robotic Process Automation (RPA): Mundane tasks, RPA tools use AI to automate, from extracting data from emails to automating entire business processes. Free up valuable time and resources for more strategic tasks, this can.
  4. AI in Marketing and HR: Personalize marketing messages, predict customer behavior, and optimize ad spend, AI can. Similarly, in HR, resumes AI can screen or schedule interviews, making the hiring process more efficient. 
Conclusion, hmm

There it is – a quick journey through the world of machine learning. Not just tech talk it is, a true game-changer redefining how we conduct business, it is. From AI chatbots to AI-powered analytics, about working smarter, not harder, it is. And the best part? A tech wizard you need not be to hop on board. As we push into a future where AI and humans work together, success they are setting themselves up for, businesses that embrace these technologies are. So, seize the day and the most of what machine learning offers, let’s make.

——————–

1980s Power Ballad 

(Verse 1)

In a world where business moves like a racing thunder,

A partner who knows, never blunders, never falls under.

Seems like a fantasy, too good, too true,

But welcome to the realm of the learning machine, it’s breaking through.

(Chorus)

Machine learning, rising like a star,

Reading patterns, near and far.

From the fifties, into today,

It’s changing the game, in a powerful way.

(Verse 2)

Data was born, and computers took flight,

In the dawn of the 21st century’s light.

Machine learning evolved, a quantum leap,

In the heart of the digital, secrets it keeps.

(Chorus)

In the world of Generative AI, boundaries are torn,

A symphony of data, a new kind of dawn.

Like an AI Picasso, an algorithmic Shakespearean dream,

Writing, composing, a creative stream.

(Bridge)

Businesses, fear not, this is not science fiction,

Harness this power, it’s our prediction.

Smooth, smart, efficient – the gear,

Machine learning, the future is here.

(Verse 3)

Chatbots speaking, with a human tone,

Analytics diving, into the unknown.

Robotic Process Automation, tasks it holds,

Marketing and HR, in the stories it unfolds.

(Chorus)

AI-powered tools, a business revolution,

Seeing through the fog, the future’s evolution.

Working smarter, not harder, the path we tread,

In this world of machine learning, we forge ahead.

(Outro)

Machine learning, it’s not just tech talk,

It’s a game-changer, where humans and AI walk.

The future’s here, it’s time to board,

Into the realm of the learning machine, we stride towards.

Let’s grasp the bull by the horns, the new day is on,

Harness the power, before it’s gone!

——————–

Jive

Groovy Intro

Dig this, picture a cat who can figure out your next big business groove, always has the right rap for your clients, and never fails to catch the happening trend. Too outta sight to believe? Not if you’re talking ’bout machine learning, man. It’s a slice of that artificial intelligence pie that’s making waves in the business world, ya dig?

A Lil’ Tale ‘Bout Machine Learning

So, how’d machine learning get its start? Well, back in the 1950s, some far-out minds figured machines could get hip to patterns. As the clock ticked on, they came up with some groovier tricks, like decision trees and a little something called backpropagation. But the real leap happened in the 21st century when we started swimming in data and our computers got some serious juice. That’s when machine learning started to really shake things up, baby.

What’s the Latest in Generative AI

Let’s jive ’bout a part of machine learning that’s really crossing lines: ‘Generative AI’. This is a flavor of machine learning that can whip up new data – like penning text that’s got a real human vibe, creating art, or even composing tunes. It’s like having a machine that’s part Picasso, part Shakespeare! Recent happenings in this field, like OpenAI’s GPT-3 and GPT-4, can grasp context, write full essays, answer questions, and you better believe it, even crack jokes.

How Your Biz Can Swing with Machine Learning

Alright, now that we’ve laid down some knowledge, let’s boogie down to the main event – how can your biz get down with machine learning? Believe it or not, it’s not as far out or space-age as you might think. There’s a handful of ways businesses, just like yours, are already getting their groove on with machine learning to make things slicker, smarter, and more efficient. Let’s riff on a few:

  1. AI Chatbots: AI-fueled chatbots can handle the customer rap, book appointments, and assist with sales. They’re a cost-effective and efficient groove for small businesses, freeing up human cats to focus on the real mind-benders.
  2. AI-Powered Analytics: Tools like Tableau and Power BI are getting hip to AI to help businesses get wise to their data. They can whip up forecasts, spot trends, and flag outliers, giving businesses the smarts to make decisions that really swing.
  3. Robotic Process Automation (RPA): RPA tools are using AI to automate the boring stuff, from pulling data from emails to automating the whole business dance. This can free up some precious time and resources for the real brainy tasks.
  4. AI in Marketing and HR: AI can personalize your marketing rap, predict customer behavior, and optimize ad spend. Similarly, in HR, AI can sift through resumes or schedule interviews, making the hiring groove more efficient.
Outro, Baby

And there it is – a quick trip through the world of machine learning. It’s not just tech jargon, it’s a real game-changer redefining how we do business. From AI chatbots to AI-powered analytics, it’s all about working smarter, not harder. And the best part? You don’t gotta be a tech guru to join the party. As we groove on into a future where AI and humans work together, businesses that embrace these technologies are setting themselves up to swing. So, let’s grab this opportunity and make the most of what machine learning has to offer, man.

 

Brian Tol is the CTO of SpinDance, a software development studio specializing in digital products, IoT and machine learning. Brian’s diverse background encompasses roles in business development, marketing, information architecture, web analytics and real-time advertising.  Away from his coding screen, Brian, a graduate of Calvin College, enjoys cooking, collecting vinyl records, discovering new board games, and making precious memories with his family.

The image accompanying this article was created using DALL-E, a generative AI technology that enables users to create new images with text-to-graphics prompts.