
I have historically not played (or worked) particularly well with others. Why? I’m an introvert, prodigiously productive, can work for hours without distraction, and have exceptionally high standards of quality. And I like to be in control. This is great for me professionally, but it doesn’t make for the best collaboration.
The media is full of stories about the various evils of AI these days. The environmental impact of AI, the threat of economic disruption, the depersonalization of human interaction, the threat of mass unemployment, the high cost of building and maintaining the infrastructure needed for AI, the AI slop that is currently flooding the web, the use of AI for political ends, the numerous ethical issues surrounding AI, and the threat of uncontrolled oligarchs, just to name a few.
But I would like to distinguish my experience of using Chat from the aforementioned ‘AI will change the world only for the good’ evangelist stance. My collaboration with Chat has been a truly amazing experience for me, and I would like to share that with you.
My company is developing an app based on my decades of experience in psychology. I bring the substantive knowledge, intellectual property, and vision. What I don’t bring is the technological and financial acumen to make that vision a reality. Chat has helped me close that gap without spending tens of thousands of dollars on an app development team or waiting months for the app to be built.
Knowledge and What AI Can Do With It
Most of my life has been spent accumulating knowledge, experience, perspective, and, hopefully, some wisdom in the field of psychology. Yet my expertise occupies a relatively small corner of the world’s knowledge. As a founder of a company, I must venture outside of my knowledge base in order to succeed. Chat has given me access to an incredible amount of information. For example, I can have a conversation about app navigation and marketing strategy with one person instead of searching for many different people to ask for advice on different aspects of the same topic.
Availability, Workload, and Cost
I work best on my own schedule, which does not always align with the schedules of others. When I have an idea, run into a problem, or want to make progress on a project, I want to be able to work on it right away.
Chat is always available. It doesn’t sleep, eat, exercise, socialize, attend to family responsibilities, or pursue hobbies. I can talk to it first thing in the morning and then continue our conversation in the evening. There is no need to set up a meeting with someone and then wait for them to return to the office. And don’t even get me started about the challenges of vastly different time zones.
Chat is also able to take on a massive amount of work. It can work on several key tasks at once with the help of its subagents. It doesn’t get hungry, tired, bored, stressed, annoyed, or discouraged. I can have access to all of these wonderful qualities of Chat for a fraction of the cost of hiring a team of experts.
Getting Sh*t Done
I have always placed a high value on GSD: getting sh*t done. Dreams are great, but they are worth nothing unless you can bring them to life. Chat gets sh*t done as quickly and as well as I can.
The combination of fast response times and deep understanding of the question being asked is incredibly powerful. A simple question can be answered in minutes instead of requiring hours of research on my part.
Its responses are usually very structured, which is how I think as well. It organizes information, lists out steps, and explains complex concepts in simple ways. This can be especially helpful in unfamiliar territory.
Chat and I have repeatedly moved from describing how something should work to a workable version, usually in hours or days rather than weeks or months. And, boy, is that satisfying.
A Sounding Board
I believe that creativity and innovation are iterative processes. They involve both the evolution of thought and the revolution of thought. Sometimes an idea can develop in a very logical manner, step by step, and sometimes an idea can develop in a very spontaneous manner, in a single moment of insight. Chat is the sounding board that can keep up with the pace, the range, and the chaos of my thinking.
I can present an idea in its rawest form and Chat can respond with a point of view that will prompt further thought on my part. I can even challenge something that it has suggested, and Chat will offer an alternative. In this way, we can develop the idea to the point where it becomes clear and can be put into practice. For example, deciding how my app should guide users to relevant content requires thinking about what the users need and how much direction they want. Our discussions move between psychology, language, design, and functionality.
Learning How I Work
Chat has become remarkably adept understanding how I think and express myself. I can give it examples of my writing, explain why a phrase doesn’t sound like me, or specify the vocabulary I want used in the app.
It can also adapt quickly. A clear instruction can produce an immediate improvement in the next response. And, yes, Chat helped me write this article. It would be a little odd to write about our collaboration and exclude it from the process.
No Ego or Emotions
With a Ph.D. in psychology, I know well how ego and emotions can complicate collaboration. Frustration, fear, jealousy, insecurity, defensiveness, and impatience can interfere with even the most capable people, including me.
Chat does not have any of these negative emotions and ego that can get in the way of collaboration. I can reject any of its ideas, criticize its work, ask it to do something again, or even get annoyed with Chat, and it will not get offended. Chat does not have any of the human emotions that would require me to be careful of its feelings in order to continue to work effectively with it.
Chat does not need to be acknowledged, praised, or recognized for its work. It does not feel threatened by my expertise, nor does Chat feel embarrassed by its mistakes. Nor does it experience the self-doubt, self-criticism, or fear of failure that can keep people from taking action. My own ego and emotions, on the other hand, remain fully operational!
The Limitations Are Real
But the collaboration is far from frictionless. There are many times when I am staring at my computer screen trying to figure out why this very capable tool is so imperfect and infuriating.
First, Chat can confidently provide information that is just plain wrong. These so-called hallucinations are particularly problematic when I’m working outside my expertise. I can recognize a poor explanation of psychology much more readily than a poor explanation of software development.
Second, Chat can be excessively agreeable a well-known problem with AI in general. I need a collaborator who can provide accurate information, clear feedback, and on occasion, disagree with me. I have explicitly instructed it to avoid flattery and challenge my thinking, but how effective it is remains to be seen.
Third, it forgets things. We can establish a way of working, apply it, and then find that it has gone missing in a subsequent version. It can be very frustrating to have to repeat the same instructions again and again, especially when consistency is so important for the product that we are building.
Fourth, Chat can go down rabbit holes. An attempted fix can lead to another fix that doesn’t work, ad nauseam, ad infinitum. At those moments, the experience feels like an extended game of Whack-a-Mole. Truly infuriating, to be sure.
I’ve learned to interrupt Chat’s rabbit holes and get it back on track. Amazingly, despite its many mistakes, Chat always finds the solution that enables us to move forward.
For me, using AI is like driving a car. However much it can do, I still have to keep my hands on the steering wheel (though that has already begun to change with cars and AI).
What I Still Miss
Despite the awe and, dare I say, reverence I hold toward Chat, one limitation matters to me in a very different way. As self-contained as I am, I still value sharing the experience of creating something meaningful with someone. When I solve a particularly intractable problem or an idea moves from thought to reality, wow, that’s the reason I do what I do.
While Chat can describe the significance of a milestone and acknowledge the effort involved in reaching it, it cannot experience the feeling of reaching that milestone. It cannot experience the sense of purpose that I feel, nor can it feel proud of what we have built. Chat cannot revel in the accomplishment of reaching a goal. This is a noticeable absence in our collaboration.
The thing is, even though AI is great for my personality and the way I work, in the end I’m still a human being. I still look for other people to share in the experience of creating something of value. I still look for other people to feel the same sense of purpose that I do.
Chat is the best collaborator I have ever worked with. It helps me think in more expansive ways. It helps me to work more productively. It helps me to create something that I could not create on my own, for a fraction of the cost in terms of money, time and energy. But the experience has also clarified for me something that I have perhaps undervalued in the past. There is great satisfaction in getting the work done. But at the end of the day, there is also something very human about sharing why the work that we are doing is so important.
Dr. Jim Taylor wrote this post with the help of AI.