BiblioTechWhat happens when you employ ChatGPT to co-write your book?
BiblioTech
What happens when you employ ChatGPT to co-write your book?
Nimrod Vromen joined CTech to discuss “Prompting Happiness: A Guide to Prompting a Life Worth Living” and how AI assisted in his passion project
Nimrod is the Founder and CEO of Ark Empowerment and Consiglieri, which aims to transform the world of professional services with AI. He is also the Chief Growth Officer for the Startup Sector at Arnon, Tadmor-Levy after 16 years as a corporate lawyer at the firm.
Recently he embarked on his latest passion project: His new book “Prompting Happiness: A Guide to Prompting a Life Worth Living” follows him as he addresses complexities in modern life. The book starts by sharing ideals from his youth and how they have changed over the years, ultimately offering a road map for navigating those intricacies and stimulating conversations on the future of AI.
Interestingly, Vromen admits in the opening pages of the book that he leaned heavily on ChatGPT to help construct and write large portions. Have we entered a new stage of art creation where humans and AI work side by side?
He joined CTech for a conversation about AI, happiness, and how to live a life with meaning in a digital world. The interview has been lightly edited for clarity.
Talk me through the idea: What 'prompted' you, so to speak, to tackle this book and this project?
I've always been dabbling in philosophy, thinking about the future, and trying to put things into structure. When I was younger, I had a diary and it had some thoughts in it. I'd written a theory for happiness that was a creative piece. The rest of my diary was a bunch of funny essays about what it meant to get locked in a toilet or what would happen if the Titanic was a German ship and it didn't sink. It was just weird ideas. Eventually, I became a corporate lawyer and for 16 years I was producing a lot of words on pieces of paper, but they were repetitive templates. I was doing all my thinking in my off time during COVID when we all got into some existential questioning.
When I suddenly started playing with ChatGPT in October 2022, my philosophizing exploded because I felt that I was playing around with a technological revolution that was akin to the atomic bomb. I felt that I had seen technological revolutions in my 17 years as a professional. I saw the Internet in 1999 as a teenager, and computers before that in 1993. I thought to myself there's been nothing like this. Every single technological revolution since the atomic bomb was an acceleration of a process that humanity has been dealing with forever, which is a process of interconnectivity working together and increased productivity.
But if you think about the atom bomb, that's a revolution where you do not live in the same world afterward. You think about AI now in that context, a day before this revolution. Humans are defined by their subjective experience of life and their collective experience of life, which is defined through their communication with one another, right? A day after AI passes the Turing test, the whole experience is going to change because now I can connect with infinite versions of humans. It's just a different world. And that set me off on a course of thoughts about my profession, about my family, about education.
Since COVID, I've been spending nights at a hotel twice every month, just to think and to strategize. And at this hotel, at 08:07 p.m., I had this epiphany: All these thoughts just put themselves into order and within 52 minutes, I had the first 70 pages. I had this process in me on how to produce them. And then from then on, it was a year and a half of iterations, which was just one of the best experiences of my life.
You start the book by sharing some of these early essays that you wrote when you were 18 on your past views and ideas of happiness. I thought it was very brave of you to publish those. Why did you choose to start the book with such a vulnerable admission?
The original book was far more academic. The original draft was about the purposes of humanity, sort of like an exposition to AI, that log of conversations with GPT and predictions for the future. And I was told by the initial readers that in areas where I exposed myself more, they connected with the book more. The readers said they wanted more on happiness and more on me because that's what tied them into the academic parts and got them engaged.
So I decided to build the book as an arc and ended on a theory for happiness in the future of AI. At which point I said I wanted to be completely candid. I'm a ‘radical openness’ sort of guy. I like to be open with people that I converse with and transact with. So why don't I just say the embarrassing thing at the very beginning, which is to share the original theory for happiness that I wrote when I was 18, and then by that time, GPT-4 came out so I could feed it with a lot of text.
If I included my original theory, I had this whole journey. Why don't I take the theory, feed the theory and the journey into GPT, and we collaborate on amending the theory for happiness 23 years later, giving it an update, like a remake?
I think for an 18-year-old it wasn't badly written. It was, there were some things that are a bit childish there, but what I liked about it was the structure. I didn't cut myself or the reader, whoever was supposed to read it. I'm not embarrassed.
One of the things that people reading this right now may not realize about “Prompting Happiness” is that a lot of the writing, especially towards the end of the book, is written by ChatGPT. So you talk about happiness and you go through your journey, and then you let the algorithm take over. Talk me through what that was like.
The first 70 pages were a pilot, and what you read were complete pages. I had the outline. I put the outline in my notes on my phone for six minutes. I went into a bathtub in the hotel, recorded myself speaking the outline for 36 minutes, and transcribed it in a transcription app for another three minutes. I went into GPT and I said, “Hey, I've been thinking about you for the last two months. You're going to change the world. And as a last act of rebellion to be true to my name, you're going to write the book about how you're going to change the world according to my outline and thoughts. Here's the text. Write the book.” And then it just wrote a lot of it.
Regarding the actual text that you see from GPT in the book, I would say it has three layers to it. First of all, I would have written all of the personal stuff and GPT would just help me edit for grammar. It wouldn't do anything else.
Having said that, there are places with an elevated layer of involvement. I am okay at writing about myself or describing something that is academic in my domain of expertise in high tech. But I don't know how to do dramatized writing. So there's a chapter where I have this argument between my inhibitions and my superego and I'm deliberating whether to do something that's borderline illegal just to release myself from stress. And I don't want to just write my logical arguments to the reader. I want the reader to see the conversation between my superego and my eternity, and so I fed all the logical arguments to GPT and I told it to write it as a conversation. So what it did there was dramatized writing. It didn't write my idea or my words, but it put it in a conversation.
Then there's one elevated layer that was responsive to my ideas and text, and that's the second-highest layer of involvement in the book. I had the framework of thoughts and before landing on the theory for happiness at the end, I wanted to give it the mic for just 20 pages, 30 pages. So I told it everything about my book that I could without telling it anything about itself. And I asked to write its own book in the same structure. And that's original, exclusively original text by ChatGPT. That's phenomenal. I hope it's in the training data for OpenAI.
You cite Yuval Noah Harari and Malcolm Gladwell as some of your inspirations - but you also mentioned how AI can create music or generate artwork. If we use AI not just to solve problems but to create art, where does the value lie in that?
Gladwell and Harari are my heroes. And when you think about the future, if you're not thinking about history, at least to see what you can learn from it, that's a problem. So they're very good at that. Regarding what I think you can do for creativity, the ideas of the planning of the creation are fully mine. I have written 300% of the text in this book.
And it was edited down to what you're reading. The AI helps me elevate the product based on more ideas and instructions that I give it. It is an actor in a movie that I'm directing. And in that sense, that's the use that I think people should use AI for. It does not hallucinate its own creative ideas unless I give it a space to do it. So that's where I think there's a good collaboration.
I want to touch on an analogy you wrote. Addressing the story of Adam and Eve in the Garden of Eden, you compare yourself as Eve, the startups as the snake, and then ‘Adam Moments’. So quickly talk me through what that is.
Startups are like bringing all these fruits from the knowledge tree, right? With all these new technologies here, it's going to take you into a new dimension. And I'm the first person who gets it from the snake because I'm their lawyer. And then I have this urge inside of me to go and convert everyone else, and specifically people who I think are just hellbound on just staying in their own box. Like Adam just wants to roam the Garden of Eden and not do anything.
So I take cab drivers when I drive back home and I just pull out a VR set and say, "Put this on, see what happens." I love to give the moments that are atom moments because what I see in the atom moment is this: On one hand, there's like profound embarrassment and God's gonna punish me and send me out of the Garden of Eden. On the other hand, there's a lot more to do in this world. There's a lot more I can do if I just let my imagination run wild. It's a good moment to have on the whole. And the thing is that history from the very first day is riddled with people experiencing Adam Moments.
Last question I have for you: what today is happiness to you?
Happiness to me is the ability to retain and sustain a positive outlook even when stress-tested with severe adversity. In that sense, I am grateful for something I would never have asked for, which is October 7, because that has stress-tested my thoughts on how to retain a positive approach to life. Just bring it in from within. You can take that approach to the grave and be sustainably content. That's happiness to me. Sustainable contentment. And that allows you to live in the moment.
All right. Thank you, Nimrod. Where can people find it to get their hands on a copy?
“Prompting Happiness” is on Amazon and you can also find it on Barnes and Noble if you're not in Israel. There is a website for it called promptinghappiness.com. The website is a community that allows others to add their own prompts to happiness, and others have added great prompts. The others will be “Prompting Harmony”, “Prompting War”, and “Prompting Love”. I want to write another three books, so this is part one of a four-part series.
