When I first encountered ChatGPT, I was astounded. As someone who witnessed the rise of mobile phones and the internet as an adult, I recall being certain that these technologies were game changers destined for success. Similarly, I saw ChatGPT and AI as the next “sure thing.” While many other “next big things,” such as IoT, Crypto, and social media, each had the potential to be transformative, their tipping points were never a certainty for me. Over the years, some succeeded, while others fell by the wayside.

Six months ago, I embarked on a project to integrate AI into every aspect of my work. I built the world’s first Agile AI chatbot, specifically designed to delve deeply into all Agile-related questions. I’ve developed training material around AI and am building a Scrum “font of all knowledge” website. I’ve used AI to tutor my son and to understand an illness. Recently, I finished two books. They aren’t selling much, but I plan to dedicate the next month to sales and marketing. Can AI be used to succeed in the sales arena? Watch this space.

Throughout all these projects, I have adhered to one rule: everything must be done with AI and automation. I use Zapier as my automation tool of choice. In terms of AI, I don’t just use ChatGPT; I use various off-the-shelf AIs like ChatGPT and Claude, and I’m building my own custom bots. I even built one to be my secondary consultant for a medical condition I have. In that time, I’ve become an expert in understanding what each AI excels at and its limitations.

AI still has many faults. For instance, I tried writing a book on motivational theory initially, but I kept being pushed down narrower and narrower avenues, each requiring more and more manual intervention. Eventually, I found myself writing the book with very little AI help, far removed from my initial goal of leveraging AI. So, I scrapped the project and pivoted to something else.

I also experimented with creating avatars, but the avatar world is not ready yet without significant investment in money, time, and expertise. Nor is AI video, although they are getting close. Another significant problem I encountered was the lack of memory in the AIs I use. They can’t maintain continuity of thought from chapter to chapter, leading to a lot of repetition in context. My breakthrough with my latest books came when I adopted a new style: the concept of practical manuals, where any repetition is less of an issue.

My first book, a comprehensive product owner manual, is now on Amazon but still suffers from some overlap and repetition. I discovered a method called thematic analysis and plan to use this to understand the overlap. I plan to spend two weeks designing an “AI-driven thematic analysis tool” to improve this hefty tome in the near future. Recently, ChatGPT-4.0 has started addressing this memory issue, so watch this space.

I do fear that the speed of AI evolution will outpace my ability to ideate and build niche products. The products I have built so far might not be needed in the near future, and this future could come very soon. I foresee books being written on the fly, based on knowledge of your skill level and depth of current knowledge. How do I stay ahead of this?

Another interesting issue is the legal stance on AI-generated content. Currently, it’s pretty much up in the air. Who owns the output from an AI? I’m not going to delve deeply into that question for the next few months. It’s not a problem unless I start selling some books!!!

One interesting note centers around my “world’s first Agile AI chatbot.” When I launched it, I reached out to my thousand plus LinkedIn connections and also offered it free to 20,000 students who took my online Scrum Master Essentials course. Yet, despite all that, the number of people who engaged with it could be counted on two hands and two feet. The interrogation of the AI never went beyond basic questions like “What is Scrum?” or “What is a Product Owner?” I was hoping they might ask something more complicated like “What are the first 10 steps to embedding the RTE role?” just to see what would happen, but nope. Why is this the case? Am I not communicating the product well? Or is it a lack of interest by people? I don’t know the answer, but if it’s the latter, that is a very important point.

I also don’t want to get into the philosophical debate: “To AI, or not to AI.” Is it wrong to write a book with AI? I have two thoughts that I keep coming back to. First, I believe the purpose of my “manuals” is to provide readers with the best possible knowledge and advice, and this is what I work towards.

Second, does AI make mistakes? I have thoroughly reviewed all the data from various AI sources and can say the output has been exceptionally good. To date, the mistakes I focus on correcting are minor, like replacing “development team” with “developers” or addressing the project-to-product debate. But the core data is impressive, comprehensive, and accurate. Sometimes, I read it and think, “This is exactly the answer I would have loved to have the skill to write.”

AI and I have developed a sort of symbiotic relationship. I walk the dog in the morning and have debates with the ChatGPT app’s avatar on topics like “the value of the Scrum Master in our teams” or the separation of the Product Owner from the Product Manager. It makes mistakes sometimes, but it’s like having an expert on hand at all times. The other day, I asked it to prep me for a role involving a niche element of work that I hadn’t thought about in years. I went into that interview confident and well-prepared.

So, there you have it. The project continues—the last six months have been transformative, thanks to AI. My main focus in the coming period will be on marketing and Sales using AI and automation but  am quite happy to chat with anyone about any of the work I have been doing.

By the way, of course, I used ChatGPT to write this. I wrote the first draft but asked it not to stray far from my, albeit flawed, style. The question is, would you be reading it if I had written it myself? Not with my handwriting!

Enjoy!

Dave

PS: here are some of my projects:

  1. Book 1: Practical Scrum Master Pocket Book
  2. Book 2: Practical Product Owner: Fundamentals
  3. My online course (not AI) :Scrum 101- A fundamental look at the Agile Scrum Master role
  4. AI-produced Scrum one-stop shop – Practical Scrum Teams (Under Construction)

https://x.com/dottykeen

https://www.youtube.com/@dottykeenan