

Discover more from It Seems to Me (ISTM)
When I first tried ChatGPT it completely freaked me out. My entire career flashed before my eyes and started envisioning a future as either n Uber driver or in the restaurant service industry.
That was only a few short months ago, months filled with enormous global debates about the benefits and/or dangers of the part of AI that we call Machine Learning (ML).
Even Geoff Hinton, the driving force and thinker that created ML has warned against its inherent dangers.
Those dangers are beside the point, or so it seems to me. AI-in all its forms-- is here to stay. There will be no laws governing it anytime soon just like there are no laws governing unpleasant practices in social media. There are social and legal issues as well as dangers surrounding many products and services that we cannot live without, cars, planes, TVs, medications, and social media. AI may contain greater dangers and benefits than all of these combined.
Transformational Event
The definitive moment came in the form of ChatGPT. Those of us who have used it
are all still learning about it. We already sense that we cannot live without it and that we have stepped into a transformational event. We speculate both on the best and worse of what AI can do and very few of us are quite sure of the answers, which will probably change anyway.
While I was fearful of ChatGPT on my first encounter, I now—just a few months later—I find I cannot live without it. It is a tool for me, not only in most of my writing but in learning about just about anything that happened before the milestone date of September 2021.
For me, ChatGPT is already a dramatic game-changer. The last time I had such an experience was in 1998 when Google Search came out, and I no longer had to go to libraries for research.
Like ML right now, there were those who steadfastly objected to Google Search, such as the Oregon teacher who was fired for banning students from using Search because she considered it to be cheating. There were also those who argued that people were too attached to horses to buy cars, and part of me wishes they were right because of global warming.
My point is that progress never stops and most miracle innovations have unwanted side effects that are sometimes destructive.
Ultimately, like Search, ChatGPT is just a tool like a hammer, and like hammers ML can be used to build or destroy Users get to choose.
My Tool & My Competitor
When ChatGPT first came out, I felt like John Henry, the fabled railroad track layer who raced and beat a new steel-driving machine, only to die of exhaustion. While I take an occasional late-afternoon nap I am far from exhausted and so far have not sensed many changes in my deal flow.
While I see ChatGPT as a competitor, I also consider it a valuable tool. Not counting the time, it takes me to fact-check it, which is mandatory, I find most writing projects including this one, get finished about one-third of the time faster. This includes books, byline articles, and white papers that ISTM reports such as this one.
The reduction in time allows me to reduce my fees to clients. But I will need more clients to meet my annual revenue goals.
Unlike John Henry, I won’t die from exhaustion and, unlike him, my competitor is also among my most valuable new tools. As a competitor, I think there are some places where ChatGPT wins and places where I am hands-down ahead. Luckily, it involves the work I love most.
If you need a deep analytical report adhering the dispassionate structure and neutral voice of a McKinsey Report, ChatGPT probably is a better choice than me. But if you want to communicate passion, vision, irony, humor, or other human traits, I think I win and I believe this will endure for the remainder of my career.
I have to admit that ChatGPT is better at storytelling and idea generation than I had imagined it would be, but it can only communicate based on historic data, so it is not yet equipped at sharing new or original thoughts.
ChatGPT’s Perspective
So, I thought it only fair to ask ChatGPT to compare and contrast me to it.
I asked, “When should someone use me to write a book for them and when should they use ChatGPT?
It replied:
“I can be a useful tool for generating ideas, brainstorming, and assisting with specific writing tasks. However human writers have the advantages of personal experiences, emotions, and deep understanding of complex topics.
Use a human writer when:
You need a strong and consistent voice or style throughout the book.
The subject matter requires personal experiences, emotions, or deep understanding that an AI may not have.
The book needs a high level of creativity, nuanced storytelling, or character development.
You require thorough research and fact-checking on a specialized or complex topic.
You want a book that reflects cultural nuances and societal context.
Use ChatGPT when:
You need help with brainstorming ideas, outlining, or structuring the book.
You're looking for assistance with generating content for specific sections or chapters.
You need inspiration or suggestions for plot points, character development, or dialogue.
You want to draft content quickly and then refine and edit it later.
You need assistance with language-related tasks, such as grammar, sentence structure, or style improvements.
It seems to me that ChatGPT and I pretty much agree.
ChatGPT vs. Me
I love the perspective and agree 100% and wonder what corner are we still not able to see around? Maybe this will seem prescient in a clearer future: https://jarche.com/2023/05/step-lively/
Great story, Bill. Thanks for sharing.