After a recent conflict with my wife, I decided to talk with chatGPT about the conflict; if you haven’t tried talking to an ai about complicated matters, you will in the near future.

For the most part, the AI wasn’t very helpful as I wanted it to be:

• Taking my side
• Having an opinion
• Gasping in shock or awe

No, none of that. And while that’s pretty disheartening when you are in the heat of your frustration, it actually was very beneficial in the long run.

ChatGPT is a language model. Kind of like the holodeck from star trek. You navigate the ethereal world of word construction with the use of a predictive language machine and can arrive at some neat insights.

And so I am slowly fizzling out of my upset as I read the AI’s uninspiring responses, it dawn’s on me that this is the first time in my life that I can have a conversation with a completely dispassionate 3rd-party perspective and arrive at an insight for behavior change.

Yes, I can read blogs. I can watch YouTube videos. I can listen to a podcast. But none of that is conversation, all of that is just entertainment.

See, humans evolved to talk through problems as a form of entertainment. In talking about problems with ourselves and with others we can process the information that is churning in our heads. That process of conversational problem solving is what has led to immense insights into human understanding (see: The Republic).

But there is a catch. When we get in conflict, which is inevitable, with those whom we solve problems with, the brain reverts to a problem making machine. It takes inputs as threats and generates outputs as violently as it deems fitting to the conflict. This can be as petty as calling someone a rude name, to slyly implying demeaning qualities about the other, and as destructive as displays of physical intimidation and violent harm. And as frustrating as that is, it is inevitable.

While that is the case, we humans have adapted to this inevitably. And we have developed different strategies and techniques for coming to some form of resolution. Because we all know that prolonged conflict is awful, and so getting back to problem solving is what has worked in the past and so it makes sense that we would work to solve the problem of the conflict. (Ironically, humans create conflicts for the opportunity to resolve them, see: game-playing).

Now imagine me sitting there talking into my phone’s keyboard transcriber all the frustrating feelings and thoughts I am having, all the beliefs and opinions I carry, and then sending that to a language model to process… Well, this is what it had to say:

Now I am not all that interested in showing you how it had said something similar to this over the course of several responses. But what I became immensely interested in was the following: a list of types of conflicts that child rearing parents have, a list of conflict resolution strategies, and a list of de-escalation tactics. Then I took those lists and had it generate a table, including a One-liner of dialogue exemplary of that type, strategy, tactic and I sat and reviewed what it had made.

The dialogue was bland and the table boring to review, but what it gave me was a moment to consider that I had many different options for coming to a resolution over this conflict. All I needed to do was consider what type of conflict it was, what strategy I might want to consider, what tactics would be useful to de-escalate the situation and finally a one-liner of dialogue to start the resolution strategy.

I must admit, I felt catharsis.

Later that night, after our son went to bed, we found ourselves seated on the couch and I was rambling on about one of my frustrations I am experiencing with one of my hobbies. And that’s when I decided to make an attempt at resolution. I leaned back slowly into the cushions, placed my hands interlocked on top of my head and said the following:

“You know, I talked to the ai at great length about our conflict this morning.”

And she took a deep breath and said pensively, “Any feedback?”

To which I replied in an even and measured tone, “In an even and measured tone, please describe the events that took place this morning.”

From there she took the time to unfurl her perspective and I made every effort to breathe slowly and deeply while imagining her perspective as if I was looking through her eyes 👀. She did great expressing her point of view evenly and in measured steps which allowed me to process and imagine her perspective. This process took time, yes, and it was difficult to stay focused on her perspective with all the mental commentary going on; but I kept breathing to an even and measured pace and was able to process her input without setting off the brain’s “Enemy:Destroy!” feature.

Once she was done, I thanked her for her perspective and reassured her that I heard her. I told her that I wasn’t going to try to feed her back her perspective, as I thought that very rude given the circumstances, but I did hear her and I was grateful she took the time to share with me in a even and measured way.

I guess it goes almost without saying that the experience was cathartic. I woke up feeling closer to my partner and I knew that I needed to share my insights with others. So I began working to build a template, a Kata, for handling conflicts more quickly in the future. You can find that 👇

And so there you have it, you now have a way to process conflicts and improve your resolution skills. I hope that my story of working with an ai was interesting, and I hope that showing how problem solving can be made easier when we work from a dispassionate language model. The Kata I provide is a language model trained specifically for conflicts with a family member, but it is conceivably possible to improve the Kata over time to include all conflicts. The only way we can do that is to get your feedback on it.

Anyway, I hope you have a great day of improving your life, even if in a really small way. Because remember, when the butterfly beats its wings, it can blow the dark clouds off from overhead.