Exclusive Analysis of ‘How Do You Use ChatGPT?’ With Logan Kilpatrick

My take on the latest episode of How Do You Use ChatGPT? with Logan Kilpatrick and the episode transcript is below for paying subscribers.

My super-secret plan is to eventually get Sam Altman on How Do You Use ChatGPT? I mean, come on—wouldn’t it be awesome to see how he uses it every day?

The first step in the plan is complete: We got our first OpenAI employee, Logan Kilpatrick, on the show. Logan’s great, and I really enjoyed getting to record with him. The main takeaway from this episode is this:

Even someone who is close to the latest and greatest language model technologies believes that they are likely to enable—rather than replace—developers over time. In the episode, Logan says that being a developer is still the most valuable skill you can have, and it’s only going to increase in value over time. He also believes that ChatGPT is going to enable the next billion developers. 

You might say: Well, of course he thinks that. He works on behalf of developers for OpenAI—it’s part of his job description. 

But I agree with him. I’ve been writing about ChatGPT as a tool to expand who gets to be a developer for a while —and I’ve seen it first hand. In the ChatGPT programming course I run, people who have never programmed before build and launch a chatbot in 30 days. They’re not incredible programmers by the end. But they have joined the ranks of people who are able to build something themselves—and that gives them the motivation to learn the fundamentals of programming that might have seemed out of reach before. 

It’s a magical thing to watch, and I think we will see ripples of it throughout the economy. Building is no longer necessarily the province of an elite few: With AI, anyone can be a builder. And that, in my opinion, is a much better world.

Timestamps

1. Introduction 00:44
2. Why learning to code is the highest leverage thing you can do 09:18
3. AI is empowering the next billion coders 13:40
4. The first prompt in GPT Builder for our video game 35:58
5. How to increase the chances for your custom GPT to go viral 39:27 
6. Prompt engineering tips while using GPT Builder 43:00
7. How to use ChatGPT in conjunction with GPT Builder 56:13
8. Ready to play our text-based strategy game! 1:06:33
9. How to finetune your custom GPT 1:19:44
10. Why you should build custom GPTs 1:43:12

Transcript

Dan Shipper (00:00:00)

Let’s brainstorm. What’s the game you want to make?

Logan Kilpatrick (00:00:02)

It would be interesting to see if we can, like, you are a leader of this civilization and then each turn, your people come to you with, here are these choices that you have to take to take the next step in the growth of this civilization.

Dan Shipper (00:00:14)

“As President Kennedy, your challenge is to balance the budget with the pressing needs of a nation on the brink of major social and technological changes. Will you prioritize space exploration, pushing NASA's budget to beat the Soviets to the moon, or focus on domestic issues, ensuring equality and prosperity at home? The choice is yours.”

Logan Kilpatrick (00:00:30)

I like that. I like that.

Dan Shipper (00:00:44)

Logan, welcome to the show. 

Logan Kilpatrick (00:00:46)

How's it going, Dan? I'm super excited about this. 

Dan Shipper (00:00:48)

I'm excited too. We've been following each other and interacting on Twitter for a long time. So it's super fun to actually get to meet you face-to-face. For people who don't know, you are a member of the developer advocacy staff at OpenAI—the first OpenAI employee we've ever had on this show. Woo! So excited. And, as part of that, you focus on enabling developers to build with the OpenAI API and with ChatGPT. And yeah, I'm psyched to have you. 

Logan Kilpatrick (00:01:19)

Yeah, this is gonna be awesome. And I'm super excited that you put this show together to really show—Part of the historical challenge of ChatGPT is showing people the art of the possible. So you doing this is making all of our lives easier. And we're also doing a bunch of stuff from a product perspective to push on this, but I'm happy that you're pushing from this angle as well. It's important work.

Dan Shipper (00:01:42)

Thank you. I really appreciate that. I love the idea of the art of the possible. I feel like that's exactly the thing that ChatGPT opens up for so many people and for me personally, so that's really great. So we were talking a little bit in the pre-production call what we want to talk about, what's most interesting, and I think one of the things that struck me is, your role is to be a developer advocate. But historically, developers have been this sort of small component of the population. And what's really interesting about ChatGPT is it's creating this world where the number of people who can become a developer, who can develop things, is expanding tremendously, both because you can actually code underlying apps with ChatGPT and because of the thing that you just launched custom GPTs, which allow you to create custom versions of ChatGPT for specific use cases. So tell us a little bit about custom GPTs and tell us why you’re  excited about them.

Logan Kilpatrick (00:02:40)

Yeah, the being able to build GPTs is probably the biggest unlock since ChatGPT itself came out. We classify everybody who's building GPTs as quote unquote builders to kind of not make people who hear the word developer are scared or say like, oh, that's not me. I don't fall into that persona. And we give this flexibility of using the GPT builder, using the custom configuration when you're creating a GPT, you don't need to know how to code at all. If you do know how to code, there's a ton of stuff that you can do, but you can actually make something that is materially much more useful than base ChatGPT itself without writing a single line of code.

Also, now there's companies like Zapier is a great example of this, where Zapier has this whole guide of how you can integrate Zapier's tools into your custom GPT without having to write any code. I think that's a huge unlock and I do think that's also the direction that we're pointing with GPTs is allowing people to do a lot more without having to write any of the code themselves today.

Dan Shipper (00:03:48)

I think that's really great. I've been using it. I have a GPT specifically for doing dream interpretations. So, when I wake up, I have a little dream, I throw it in there and it just gives me a little bit of analysis. And then it does a picture with DALL-E and it's really great.

And I use it a lot. But I have a bunch of other ones. I've been working this weekend on one, where I want to be able to do a chat with it and then have it save the notes from it from my chat into Notion. And so I'm hooking up to the Notion API. I got a little stuck on it, but one of the things that I think is really cool, you talked about the GPT builder and that is a GPT that lets you build GPTs So it's sort of GPT is all the way down. And I think that's a really important thing for people to notice and realize is that as you start stacking AI on top of each other, you get extra leverage where now you have a GPT builder that lets you build GPTs and that lets you build more things more quickly than you would ordinarily.

Logan Kilpatrick (00:04:56)

Yeah, and it's interesting because when you use the GPT builder, it doesn't feel like you're using a GPT, which is interesting. So it's kind of cool that, as you abstract out of the interface where people are very aware that this is some specific tool, you also kind of lose the fact that it's AI all the way down. It feels like some other system that we built, which I think speaks to the general reasoning engine capabilities that large language models have, which is really interesting.

Dan Shipper (00:05:23)

Yeah, that's really cool. So, we're going to get into, we're going to build something together. We're going to build a custom GPT together, which I think would be super fun. Before we do that, we always like to talk about historical use cases or just ways that you use ChatGPT yourself. And I know there's another custom GPT you might want to talk about, but, generally, tell us how you are using ChatGPT and custom GPTs in your own life and in your own work as a place to get started.

Logan Kilpatrick (00:05:55)

Yeah. I think the biggest, highest-leverage thing for me as somebody who's still doing engineering work on a daily basis is coding stuff. I know there's a ton of really useful GPTs around coding and making stuff. Most of them have been historically focused on low-code type of, give me a single line prompt and I'll go and build you this thing. And that's not the use case that— I'm more around I'm not a Next.js expert and I'm in a Next.js codebase or I'm going to typescript code base and I'm trying to build something or add a new feature and I have a computer science background, but I'm not formally trained as a web developer. I never did that as my job. So I'm always asking ChatGPT the really, really dumb questions. And you'd be surprised the number of features that are exist on our API platform that were built because ChatGPT helped me—the feedback, thumbs up-thumbs down is a great example of that, when there's a outage in the API and there's a little pop-up that shows up dynamically, all of that was built by ChatGPT itself, which is really cool.

Dan Shipper (00:06:57)

Yeah, I think that makes a lot of sense and it sort of dovetails with the thing that I've been writing about or thinking about a lot is: ChatGPT's facility at summarizing, and once you start to see how good it is at summarizing, you see summaries everywhere in your life and in everything that you do. And I think that that is true also of developing where people would be surprised, if you're not a developer, how often being a developer is just summarizing the API docs or summarizing a Stack Overflow answer into your, into your codebase or whatever. And it sounds like that's a lot of what you're doing. There are some parts of programming that really, you're doing fundamentally new stuff that no one's ever really done before. And in that case ChatGPT might be sort of helpful to think it through or whatever, but it's not going to be as helpful. But most of developing is really just stacking bricks that have already been laid in a slightly different configuration than they've been laid before for your specific use case. And that's where I think it shines so much because it's so good at knowing what people have done in the past and just writing a little snippet for your specific scenario. 

Logan Kilpatrick (00:08:02)

Yeah, it also has been really helpful to me as I look at 300 lines of code and not— I kind of understand, I read the code that's right around the thing that I'm trying to make an addition to our modification, but I don't in my personal context window, I don't have all 300 lines of code. And it's really helpful when I then go and bring that into ChatGPT and say, here's all the code, ChatGPT now actually has all 300 lines of that code in its context window and it can more effectively solve the problem, honestly, than I can, because it's really difficult to take 300 new lines of code, understand it, and actually conceptualize how you can make some modification to it. 

Dan Shipper (00:08:44)

That makes sense. And I guess I have to ask, cause you're in this really interesting position where you're at OpenAI, you're watching all the progress, you have access to internal stuff, you're a developer yourself. How is using ChatGPT, making custom GPTs, seeing the progress internally at OpenAI, how is that changing what you think the future of being a developer is? Is learning to code still a thing? How are you thinking about this over the next year, five years, all that stuff? 

Logan Kilpatrick (00:09:18)

Yeah, learning to code is still definitely a thing. I think if you're looking for the highest-leverage thing that you can do in your life, learn to code. I think this will continue to be the case in the next five-to-ten years as well. My expectation is that folks will be able to solve probably 10-to-20x and it's going to continue up this exponential curve of the difficulty of how conceptually hard the products and the services are that you want to build just by yourself. To me today, I feel like if you had taken me back two years ago, I feel like I can solve 10x more ambitious problems. I have some idea in my head. I want to build a restaurant application. I can actually go and build that myself and two years ago, I would have been like, that's probably a little bit too much. Maybe I'll go hire someone to help me do that. And, I think continuing up that exponential curve, soon I'm going to be able to build an entire empire of products and services for my company just by myself. And I don't think three years ago people would not have imagined they can do that.

And that's why I still think developing today is one of the highest-leverage things that you can do, but in two to three years and five years. It is going to be even higher leverage and the gap between people who can who can code and who can't code is going to continue to grow and I think the nice part is AI can help bridge that gap but there will always be limitations on the no-code tools as far as what you can build you need to understand how the systems work.

Dan Shipper (00:10:53)

Yeah, that makes perfect sense. I mean, you're speaking my language here. Cause I'm running a small company where I write a lot, I do courses, and we develop software products and it's a small, profitable lean team. And the thing that I love so much about tools like this is, we can go so far with the small group of smart people that we have, because now we all have ChatGPT. And I think you're totally right. It lowers the bar to getting started on something so much that for me—a project that I never would have attempted before. Cause it would be like, I only have a couple hours here. I can suddenly do it and it's out in the world. And that is the coolest thing ever.

I think that's one thing that people don't quite understand. I actually talked about this a little bit with— I had Tyler Cowen on the podcast last week. And I actually talked about this a little with him where, for really skilled people, it can, it can increase your productivity where you're slightly faster, but it also allows you to do things that you never would have done before because the bar would have been too high to get started. There's an activation energy to making projects. And I think that that's incredibly cool. And I think you're also right. It does something where if you can't code at all, you can suddenly make stuff and it'll be better if you understood the underlying technology and that'll work. But the really cool thing is, I teach this chatbot class where people learn how to build a chatbot. It's not with GPTs. It's the actual underlying, make the whole thing a web app—all that kind of stuff. But you use Chat to help you code and it takes people from, having never developed in their life to, I built and shipped an app in 30 days. And it's so cool to watch them be able to make stuff because they all have ideas, but they've never been able to put them out in the world before. And Chat becomes this thing that enables them to do it. And also it gives them a lot of motivation to learn the underlying fundamentals because previously, if you wanted to learn to code, you spend six months, learning if statements and for loops and all this stuff  where you're like, I have no idea how this levels-up into the app I want to build. And what ChatGPT does is day one, you make the thing that you want to make and then day two, you run into some bugs and you're like, well, I guess I have to figure out what an if statement is. But you now have a lot of motivation to do it. And ChatGPT can explain it to you and it’s just really, really cool. 

Logan Kilpatrick (00:13:35)

That resonates with me so much. After we're done recording this, I'm going to have to get the clip out and play it back to myself every month or something because this technology is what is going to be the thing, the catalyst to enable the next billion people to learn to code.

There were not enough computer science educators and resources on the planet to teach the next billion people to code. It was not going to happen. And I think it's so crazy to think that because we have LLMs, people are actually going to be empowered to go and do this in a way that's personalized and empowering to them specifically. That might be the real thing that ends up being the most beneficial of this technology, of all the things that it can do.

Dan Shipper (00:14:15)

I think so. Yeah. It expands access to being able to build significantly. And I think that's the coolest thing. 

Logan Kilpatrick (00:14:26)

It's beautiful.

Dan Shipper (00:14:28)

Yeah. And with that, maybe we can get into some building. But before we start actually building something together, which we're going to do, I think you wanted to show me a custom GPT just to sort of whet our appetites. Give us a GPT that you built. Tell us about why you built it and what it is and why you think it's interesting to look at.

Logan Kilpatrick (00:14:44)

Yeah, 100 percent. Let me get my screen-share pulled up and then we'll, alright. So we're in the GPT store. So I was lucky enough to be able to build Planty, which is a little play on words. If folks watch the OpenAI DevDay back in November, there was a customer video of people talking about how they were using ChatGPT, how it's impacted their life. And there was this incredibly cute old man who came on, with a really grainy video sitting in his office. And he was talking about how Chatty has sort of unlocked all these use cases. And he was referring to ChatGPT as Chatty, and it was really cool to see him get that value and have that relationship with this tool.

So Planty is a little bit of a play on words with that. And, the whole concept was, literally sitting on the couch with my girlfriend and we were thinking of, what are the cool things that we should build for people? And she's not somebody who's in AI, but she's somebody who loves house plants and gardening and everything like that. The first thing out of her mouth—so all credit goes to her—was we should build Planty. And it's really designed as a houseplant caretaker. And she has a Monstera and she needed to do some stuff with the roots. That was a little bit over my head from a plant technical perspective.

And she went in, as we were building Planty, used it to actually solve that problem. I got my older brother for Christmas, a lemon tree. And as soon as it got into our apartment, all the leaves were gone. It looks very desolate. So, I put out a tweet about how if Planty can save this lemon tree, I'm a deep believer in AI. And, really, you can do things like just ask simple questions, or get really tactical and upload images and be like, here's the color of the leaf of this plant. Why is it like that? And it really unlocks, again, ChatGPT could do some of these things, but providing a bunch of guardrails to allow it to be able to do this. And, giving it this friendly personal feel. I think this actually brings to life what's possible with this GPT.

Dan Shipper (00:17:07)

Tell me about that. How would I know whether or not I should make a GPT versus just vanilla ChatGPT?

Logan Kilpatrick (00:17:16)

That's a great question. I think that this is one of the tricky trade-offs today. I think what will end up over time being the way you answer this question is like social signals. I think people will probably come up with a bunch of—if you have a plant question and there's like 100,000 people who said that Planty was the thing that solved their problem, you'll probably just go there because that's what it seems like you should socially do. I do think the interesting thing for people to keep in mind is, behind all the scenes, behind everything, it is inherently ChatGPT that is still answering these questions. So it's not some net-new capability.

I think the thing that we did with Planty is like grounding it in very specific stuff, like if it needs to go and look something up, it goes to a bunch of specific websites that we pre-vetted as places of authority that you should go to instead of, I don't know, we want to sell you plantstuff.com, which might have a biased perspective. So there are a few nuances to how it's being built, but again, fundamentally, it really is just ChatGPT behind the scenes with like a few layers of abstraction on top of it.

Dan Shipper (00:18:27)

That's really cool. So I actually have a bunch of plants. As you can probably see, I have a fiddle-leaf fig, in my apartment. And I actually do have a question randomly, that might be worth doing. I don't see your screen anymore.

Logan Kilpatrick (00:18:42)

Yeah, I just switched it to Planty.

Logan Kilpatrick (00:18:50)

I’d love to see the question. Let’s do it.I ’m curious what Planty's suggestion is. 

Dan Shipper (00:18:52)

Here's the thing. I have this fiddle-leaf fig. Fiddle-leafs are supposedly very finicky and hard to take care of, but I've had it for like four years or five years or six years or something like that. And I just water it every week and a half or something. And it’s doing pretty well. Every once in a while, a leaf will fall off and I freak out, but then it's fine. But one of the things that I don't know is, I think you're supposed to repot these things at some point, but I just don't know when to do that or how to do that. And I'm just wondering if I'm like artificially constraining my plant and I should be repotting it, and how I would know. 

Logan Kilpatrick (00:19:39)

You said it’s a fiddle-leaf fig? How and when should I repot my fiddle-leaf fig? 

Dan Shipper (00:19:41)

So you correct your typos in chat cause I never correct my typos. I just type like a maniac. Is that a bad thing? 

Logan Kilpatrick (00:19:43)

I do type like my maniac, but it's more of a habit thing where like if I stop correcting my typos in this, then I'll stop at everything else. But people will be pissed at me because I'm typing like a maniac on Slack or whatever it is.

Dan Shipper (00:19:58)

That's so funny. It's just funny doing this show. Cause I'm typing stuff a lot in screen-shares that get recorded and broadcast to thousands of people. And I'm just like it looks like I'm in third grade or whatever. And I'm like I swear I can actually type. I just don't do that for ChatGPT.

Logan Kilpatrick (00:20:20)

Yeah. I'm chronically a bad speller, so I need to do it. So I asked the question of how and when should I repot my fiddle-leaf fig. What are the trade-offs from not repotting and, I think, touching on the artificially constraining it. So it's interesting. “When to repot.”

Dan Shipper (00:20:42)

So it said “Repotting your fiddle-leaf fig can be quite the adventure.” Is that part of your personality you gave it? 

Logan Kilpatrick (00:20:51)

Yeah, it is. I think it's supposed to be this friendly, exciting thing that's helping you instead of—I think by default, ChatGPT is pretty generic about just trying to give you this information. So I think it brings it to life a little bit.

Dan Shipper (00:20:56)

Okay, cool. And if you scroll up, it says, “Repot your fiddle-leaf, ideally during the growing season, which is spring or early summer. Take a peek at the drainage hole at the bottom of the pot. If you see roots starting to circle around or peek through, it's time for a bigger home.” I mean, I'm going to just check that real quick. Give me one second. We'll see how it goes. I'm super curious. See roots starting to circle around or peek through. Hold on. There are literally roots coming out of the side of the pot. It's not peeking through. It's like they've grown out and they're coming out basically almost onto the floor. So I think I need to repot.

Logan Kilpatrick (00:21:49)

“How to repot it.” This is perfect.

Dan Shipper (00:21:50)

Yeah. “Find a pot that's about two or three inches larger in diameter than the current one. Make sure it has some good drainage holes. Add some fresh potting to the bottom of the new pot. Fiddle-leaf figs like well-draining. Gently remove the plant. Check the roots and place your fiddle-leaf in a new pot and fill around it with more.” What I want to know—So here's my last question: I'm super lazy and I am not going to figure out the diameter of the pot and do all this stuff. Is there some way I can get someone to do this for me? 

Logan Kilpatrick (00:22:40)

Where do you live?

Dan Shipper (00:22:42)

I'm in Fort Greene, Brooklyn.

Logan Kilpatrick (00:22:48)

Fort Greene, you said? See, there's my recommendations typo—horrible word to type. 

Dan Shipper (00:23:02)

So you said, “Are there services or ways I can get someone to do this for me? Any specific recommendations for someone in Fort Greene?” And Planty says, “Absolutely. Getting some help with repotting your plants is a great idea, especially if you're not up for a DIY project or if your fiddle-leaf is getting a bit too hefty to handle on your own.”

Definitely too hefty to handle for me. “In Fort Greene, you're in luck because it's a vibrant area.” Definitely. Thank you, Planty. “With lots of plant loving folks and services.” And it gave some, I would say generic recommendations, like local nurseries and garden centers, although it did give us a couple of specifics within that saying Natty Garden or Greenery Unlimited.

Logan Kilpatrick (00:23:38)

Are those real places? 

Dan Shipper (00:23:40)

I don't know. I'm not up on the local plant caretaking services in Fort Greene, but I would love for you to check it out. It also recommended task websites like Taskrabbit or Thumbtack, or plant care services like The Sill, which I definitely know is real. A friend of mine's wife runs that company. So, definitely got some good directions. Maybe, after this episode is over, I will get it taken care of and we'll do a part two where I update you on the progress of my fiddle-leaf. So it looks like you looked it up and Natty Garden is a real place.

Logan Kilpatrick (00:24:20)

It’s a real place. It looks nice too. I kind of want to go to Natty Garden. This looks like it would be a good time regardless of if you get your plant repotted there or not.

Dan Shipper (00:24:31)

Great. Well, I will be checking out Natty Garden after the episode. And I'll send you pictures. 

Logan Kilpatrick (00:24:37)

This episode is not sponsored by Natty Garden, but it looks like an incredible local community store.

Dan Shipper (00:24:42)

Cool. That's great. I love Planty. I think it’s a cool little idea. And, I think my fiddle-leaf is probably pretty happy that you showed that to me as well.

But, now I want to move us on to the main event. So, I told you that one of the really fun things about this show is getting to make something together or do something with ChatGPT where we’re  both sort of like, is this going to work? And that's sort of the magic of it. So you proposed making a game in the GPT builder. And I wanted to start with why did you propose that? Why do you think that's exciting?

Logan Kilpatrick (00:25:22)

Yeah, I think if you were to think about how you would have had to do something like this even just pre-GPTs, just with ChatGPT and plugins, I've been brainstorming this idea for a while of, I would love to take people through if you played Civilization or Empire Earth or any of those games that are situation-based where you're building an empire over time.

Those were always fun to me. And, it really brings history to life, especially if you can sprinkle in actual facts about real things that happened. And without GPTs, it was impossible to do this. I, as the developer or user, would have had to pay to generate these images. Each individual image, it would have cost me money. The constraints of building it's actually kind of hard to do. And it's a non-trivial amount of work. I think the cool thing now with GPT is you can use the GPT builder. You can sort of have it take care of a lot of the nuances of game design.

And, we build in DALL-E. We can use built-in browsing to pull in facts in real time, keep track of things for you, potentially use Code Interpreter to do some actual simulations of, I dunno if we can simulate battles or like statistical likelihoods of things, but it's incredible what will be possible with this. So I'm really excited. And also it's one of the areas that I think specifically because there's not a category right now. It's a little bit underexplored. If you look at the trending categories in the GPT store, there's not really a ton of games that are in there, if any games at all.

Dan Shipper (00:27:07)

So I think it's an opportunity for us. That's really interesting. I wonder why, cause I know, in the '80s and '70s, during the first PC revolution, all text-based games were super, super popular. And it feels like an obvious thing to try. I wonder what's holding people back. Have you seen any at all?

Logan Kilpatrick (00:27:24)

I have seen a few. So actually the most interesting one was this game, which was one of the most popular plugins, and it was called Algorithma. And it was this guy and his and his daughter, I'm pretty sure, the daughter learned to code to build Algorithma and ended up scaling it to the point they had like 5 million unique users or something like that, or 5 million like users over the course of a few months who used Algorithma. And it's actually a little kid who built this thing. And, it was actually fun to play. It was crazy. And I talked to the dad and he was saying how she was using AI to help her build it. So it's a great example of the theme of people being empowered to build cool things. Yeah, I think that was probably the biggest one that I haven't seen their GPT, but I'm guessing it's doing a bunch of really cool stuff.

Dan Shipper (00:28:16)

That's awesome. So opportunity, some white space for us. Maybe this is the next big thing. Maybe we'll do it live. I guess to start, let's brainstorm. What's the game you want to make? If you have a couple of ideas, throw it out and I can help refine.

Logan Kilpatrick (00:28:33)

Yeah, I think it would be cool to—and I think this is, maybe the decision for us to decide is do we want it to give it a purely fun and exciting type of vibe or is there some educational component? Would we want people to be able to learn about history as they go through this? I think there's a trade off either way. In some sense, people just want to escape and play some fun game, but there could be some value— And, maybe we can have a toggle where like we let people at the beginning of the game decide are you here for fun? Are you here just to actually learn about stuff as well as you go through?

Dan Shipper (00:29:14)

Here's my feeling. I think the core of a game or really most content has to be about fun. It's about creating an amusing experience. And if you happen to learn something along the way, I think that's amazing. This podcast is sort of similar. Ideally, it’s super fun to listen to. And you happen to learn something along the way, but we don't want to jam in the educational stuff  because then that gets boring. So but I think what I hear you saying is history is kind of an interesting thing for you and having a part use history in some way, it would be kind of fun. Do you have any specific ideas for how to incorporate history into a game like this?

Logan Kilpatrick (00:30:03)

Yeah, I'm curious if the idea of—and maybe history is too dark for this and it won't work, but I was playing as different world leaders or something like that. And maybe we need to make up something so that we avoid all the darkness because history is a very dark thing. Maybe that's the route that we want to go so that we avoid, like I was thinking of Julius Caesar or something like that. I'm sure Julius Caesar's done a bunch of horrible things and we might want to try to recreate that in a game. But it would be interesting to see if we can follow, like you are a leader of this civilization and like you get to choose, and then each game turn is like your advisors or like your people come to you as here are these choices that you have to take to take the next step in this, in the growth of this civilization.

Dan Shipper (00:31:01)

And I think that's great. I love that. And it would be fun. It's like, maybe you can pick any, like a couple of big historical turning points or events and see if you can do it better. It's like, you're, you're JFK deciding about the Bay of Pigs, whether to do the invasion and each turn, someone's coming to you and you have to make some decisions and you have to make all the decisions to see if you can make the Bay of Pigs—either just don’t do it or make it work out or something like that. Maybe we could pick a couple historic events where it seems like things went wrong. And it's like, can you do it better? Do you think you could do better than JFK?

Logan Kilpatrick (00:31:46)

I like that. I like that. I think the key for me is finding things that are uncontroversial enough, or ones that we both have enough information where I just have so little context on the Bay of Pigs, other than that it happened where I don’t know if I—

Dan Shipper (00:31:54)

Me neither. I'm sort of making that up, but yeah, let's find something that feels like it would be easy to do or not easy to do, but find something that feels like it would be good. And not get too dark.

Logan Kilpatrick (00:32:11)

Yeah, I'm curious if we can do—

Dan Shipper (00:32:13)

This is a good thing to ask ChatGPT about, by the way.

Logan Kilpatrick (00:32:15)

Yeah it is. I think the two that come to mind that would be really interesting is potentially some of the land purchases that the U.S. made back in our early founding history of play out the scenario of the world if we don't make certain land purchases. And, how does that structure, if we're sharing North America with other people or potentially the—I don't know if you've seen the show on Apple TV about what happens if the Russians beat the U.S. to the moon. But I think that could be an interesting way of playing out that scenario as well.

Dan Shipper (00:32:54)

I like that. So I think land purchases are a little bit colonial for me. 

Logan Kilpatrick (00:32:56)

Yeah. I'm with that. I'm with that. 

Dan Shipper (00:33:01)

It's giving imperialism. But the moon is kind of cool. I think the interesting thing is I feel like the way this game probably has to work is like you're doing—I mean, I guess we could say like, you're the director of NASA and, before the Apollo missions were the Gemini missions and Gemini, I think one of the later ones blew up and a bunch of people died and it was like, are we going to keep going? And maybe you're the director of NASA and you have to figure out how to get the Apollo program off the ground and get us to the moon or something like that. It could be kind of fun, but I really think we should be refining in ChatGPT. 

Logan Kilpatrick (00:34:00)

Yeah, let’s do it. Let’s go in and start building. I was also thinking there's probably a separate game, which maybe we don't make in the context of this, but around deciding how the U.S. spends their budget every year and how potentially the different allocations lead to some very different outcomes.

Dan Shipper (00:34:21)

That’s really good. I love that. Would you rather do that? Which one are you more excited about? 

Logan Kilpatrick (00:34:26)

Well, I'm happy for either. I feel like the U.S. budget one, who knows, maybe it could be very optimistic and very positive if we don't spend a ton of stuff on military budget, but I'm not sure. I feel like, yeah, potentially controversial, potentially not. I'm happy either way.

Dan Shipper (00:34:56)

Alright. You're going to need to make a call here. Cause I want to do whatever is exciting to you. So tell me which one—we're going to make you make a choice. Tell me which one you want to do. 

Logan Kilpatrick (00:35:08)

Let's do: You’re president and you get to sort of choose the time period or choose the president. And then we'll have ChatGPT dynamically pull up like the budget of the U.S. government and the allocation of funds, and then let you sort of reallocate and then like play out American history potentially, with different allocations of budgets and see how that affects the outcome of the world.

Dan Shipper (00:35:32)

I love it. I love that. Should we throw it into ChatGPT and have ChatGPT refine it? Or do you want to go right into the building configuration-type thing? 

Logan Kilpatrick (00:35:50)

Let's use the GPT builder and see if it's—I feel like I've been using this recently and it's been helping me get to the point of having something that's really interesting that we can always jump out and go into a different window and have it do something a little bit more verbose for us if we need to.

Dan Shipper (00:35:58)

So we are here in the GPT builder. We have created a new GPT and we’ve said, “We want to make a game. The core concept is that you get to choose which president you wanna be, and then the GPT will go and search the U.S. government budget slash distribute spending in that budget. Let the user reallocate the budget and then play out the world with that updated allocation.”

And, one of the things that I think it's interesting about this prompt that you gave me is we're not asking it to help us refine specifically. We're just letting it go straight ahead. And I think that that has to do with some nuances about how the GPT builder works and its instructions. I'm curious if you want to tell us really briefly about that.

Logan Kilpatrick (00:36:35)

Yeah, the GPT builder behind the scenes—and folks can see we've sort of opened the source instructions for the GPT builder itself, but it is doing a little bit of this refinement. We could probably and maybe we should try this and see if we can iterate on it once we've actually gotten what it outputted. But it does do a little bit of that prompt-engineering magic behind the scenes for us, which is nice so that you don't have to spend time doing that.

Dan Shipper (00:37:05)

I see. Cool. So if I went into the configure page, I would see the instructions that it wrote and it says, “The GPT's role is to assist in creating a game where players choose to be a U.S. president and reallocate the government budget. Based on the user's allocation, the GPT will simulate and describe the outcomes of these financial decisions. It should access and interpret data from the U.S. government's budget and spending, providing insights and hypothetical scenarios, the GPT must ensure accurate representation of data and responsible handling of historical and political content.” So what you're saying is it sort of rewrote it in an interesting way to create a game beyond what we said in our initial prompt, which is kind of interesting. And so now it says, “To start shaping this GPT, let's give it a name that reflects its unique role. How about Budget Commander?” What do you think? Budget Commander? Should we keep going? How are you feeling about it?

Logan Kilpatrick (00:37:52)

Yeah, I like a good one-word. If it can come up with something that's like, I don't know if it'll be able to do that or not, but—

Dan Shipper (00:38:05)

I'm saying I like a good one-word name. Can you find one for me? 

Logan Kilpatrick (00:38:09)

Something not boring, hopefully. We'll see. Budget Commanders a little—

Dan Shipper (00:38:14)

Allocato. 

Logan Kilpatrick (00:38:17)

Oh, I like Allocato. That's not horrible. That is not horrible. 

Dan Shipper (00:38:21)

What about Allocator? Could be kind of interesting because I feel like Allocato is not necessarily—I don't think that's a real word. But the really interesting thing is, I just published this piece about how the knowledge economy is over and the allocation economy is beginning, where in a knowledge economy, your economic value is sort of pegged to what you know and being able to bring to bear you know to the task at hand. And in an AI world, it's a little bit more about knowing how to allocate resources, because in a lot of ways you're abstracting yourself out of certain day-to-day tasks. And so knowing how to allocate the resources of intelligence to do those tasks and how to edit them, or whatever, you become in a lot of ways a manager. And so you're learning how to allocate resources. So I kind of like Allocator because it's an unknowing reference to something that I've been thinking about a lot. So, I'm saying, “I like the name Allocator. Let's go with that.”

Logan Kilpatrick (00:39:27)

The thing to keep in mind as you're building GPTs is, and I don't think Allocator is going to be one of the ones that ends up having conflicts, but you really do want to, if you're trying to build something that a lot of people are going to use, you want to kind of own the SEO space. And the more generic that you make the name, the more polluted it'll be as somebody is trying to find it. So it's just like a general tip as you're thinking about this is you want people to be able to find what you're building. So name it something such that people will actually be able to find it.

Dan Shipper (00:39:58)

That makes sense. That makes sense. So it's given us a profile picture for Allocator. It says it's designed in a futuristic sci-fi style and, yeah, it's sort of abstract futuristic. He looks like there's technology and metal and lasers going on, but I don't think—

Logan Kilpatrick (00:40:17)

That’s not the vibe.

Dan Shipper (00:40:20)

I don’t think it’s quite the vibe because we're talking about historical stuff and a president allocating resources. So maybe I'll say, “I don't think that's the vibe. This game is about historical events and the president allocating resources. Can we do something more like that?”

Logan Kilpatrick (00:40:45)

I would add something about the simulation perhaps as well because I'm wondering if that will—I feel like I feel like what we're gonna get now is the president's sitting on like a stack of gold coins or something like that, which maybe will be okay, maybe that's what we want, but I'm curious to see okay, if it takes it. Oh, interesting.

Dan Shipper (00:41:02)

Interesting. Yeah…

Logan Kilpatrick (00:41:06)

Yeah, this profile picture looks like, imagine the Declaration of Independence with a big presidential stamp seal right in the middle of it is what it looks like. And a bunch of really, really tiny written text around the sides. 

Dan Shipper (00:41:20)

It's giving video game made from that Nicolas Cage movie.

Logan Kilpatrick (00:41:26)

Yeah, I forgot. Oh, National Treasure. I love National Treasure.

Dan Shipper (00:41:28)

Yeah, it’s the custom GPT for National Treasure. 

Logan Kilpatrick (00:41:35)

That's a great example. 

Dan Shipper (00:41:38)

What do you think? Should we keep going? Should we add simulation? Or are we good with National Treasure as a jumping off point? 

Logan Kilpatrick (00:41:46)

Maybe we take National Treasure as a jumping off point just to see if we can borrow Nicolas Cage's brand and get him involved in some capacity. Maybe the new framing is we bring Nicolas Cage into this as the Easter egg.

Dan Shipper (00:42:02)

Yeah, or Nicolas Cage is like, when the DALL-E does the drawing, Nicolas Cage is the president. That could be pretty good. Yeah. If we don't run into content restrictions, that is. 

Logan Kilpatrick (00:42:13)

That is going to cause content restrictions.

Dan Shipper (00:42:17)

Okay. I’ll say, “Cool. This looks good. What's next?” Is that okay? How do you feel about that? So basically I'm going to accept our National Treasure-looking image and now it's talking, it's asking me. “Now let's refine the behavior of Allocator to ensure a line that aligns with your vision for the game. We'll start by discussing its role and goals. Could you tell me more about how you envision Allocator interacting with users? For example, should it focus more on providing historical facts or on creating immersive hypothetical scenarios based on budget decisions?” 

Logan Kilpatrick (00:42:55)

I feel like it needs to do both. And I think the thing that most people are going to struggle with is there needs to be some foundational core mechanics of this game. And, I'm not somebody who knows how to build games and what those core mechanics could be. So I wonder if we should tell the GPT builder that we need it as the expert on or with its expert knowledge on how to build extremely fun and engaging scenario-based games to leverage that expertise to come up with a framework for this Allocator game.

Dan Shipper (00:43:31)

Okay, so what I'm writing is, “I feel like it needs to do both. There needs to be some kind of foundational core mechanics of this game, and I'm not really sure what those should be. We're not experts. We need your expert knowledge on how to build extremely fun and engaging scenario-based games to come up with a framework for the actual gameplay mechanic.” And the thing I want to point out is one of the things that you said is like, “We need your expert knowledge on how to build extremely fun and engaging scenario-based games.” Tell me about why you included that. 

Logan Kilpatrick (00:43:58)

I think it's literally about empowering the GPT and, in this case, the GPT builder to have agency that it knows how to solve this problem and leverage its domain-specific—by default, the answer that you'll get would be  very basic and perhaps not following the best practices once you sort of influence it to know that it's an expert in that field is one of the prompt-engineering techniques that folks often recommend.

Dan Shipper (00:44:27)

Yeah, definitely—making it more specific, asking you to simulate being an expert. I'm curious, do you have any specific scenario-based games that you're thinking of that we could even get even more into that you think could be inspirations for this?

Logan Kilpatrick (00:44:39)

Yeah, that's a good question. I mean, I think the two games and they're not scenario-based, but they're like the closest example to this is I played Civilization growing up and I played Age of Empires growing up and those are probably the two examples, but those are a little bit like different games where maybe we could take inspiration from that.

Dan Shipper (00:44:58)

Okay. So I'm adding, “A couple of games that are inspiring to us are Civilization, which if I could spell it—and, I'm falling into correcting my spelling mistakes here, the GPT builder. I'm less comfortable making errors because I don't know what it'll do. “Inspiring us to us are Civilization and Age of Empires. They aren't exactly the same as the game we're making, but the vibes are an inspiration.”

Logan Kilpatrick (00:45:30)

It's all about vibes with these GPTs. 

Dan Shipper (00:45:32)

All about vibes. Okay, I'm gonna hit enter. Are we good with that? 

Logan Kilpatrick (00:45:34)

Yeah, let's see it. 

Dan Shipper (00:45:38)

Let's see what happens.

Logan Kilpatrick (00:45:44)

I've got a good feeling. 

Dan Shipper (00:45:49)

Yeah, this is going to be the best Nicolas Cage video game ever created. 

Logan Kilpatrick (00:45:52)

Nicolas Cage needs to be the assistant that comes—your trusted advisor. Whoever the president is, it's Nicolas Cage in all of these, he comes in out of nowhere.

Dan Shipper (00:46:04)

Or in order to avoid content restrictions, it's like Kicholas Nage, you know?

Okay, so it said, “Now that we have a clearer idea of Allocator's role, let's think about what should be emphasized or avoided in its interactions. For instance, are there specific historical periods or presidents you want to focus on? Also, how should Allocator handle sensitive topics that might arise from political discussions?”

So, I think we were expecting a bit of back and forth with it. Here's what I think it should be, here's the mechanic, but it sort of just plowed ahead. I'm curious how you would handle this. 

Logan Kilpatrick (00:46:41)

Yeah, the reason why it's it's just plowing ahead and, perhaps there's not enough nuance and the instructions that it has, but it's really following this, step one. And then at the end of each step, there's literally a command that says go to step two. So I think we probably would need to tweak those instructions to be like, hey, maybe give the user some input and then let them go back and forth on some of these things. Otherwise, as it stands today, it literally is just going to keep plowing. And I'm curious if we can go to configure, if we can see what it added. 

Dan Shipper (00:47:15)

Cool. So we're going to click configure. And configure is sort of like we're in this chat environment where we're creating it, but configure, you can kind of look behind the hood and see what it actually made.

The instructions now say, “Allocator’s role is to plan historical accuracy with engaging hypothetical scenarios in a game focused on U.S. presidential budget allocation,” which is funny because I don't think the president allocates the budget. I mean, does the president allocate the budget? Congress definitely has to at least approve the budget. So, “Drawing inspiration from games like Civilization and Age of Empires, Allocator should offer a strategic scenario-based gameplay experience. The core mechanics will involve users selecting a U.S. president, being presented within the actual budget of that time, and then reallocating funds. Allocator will then simulate the potential outcomes of these decisions, offering both factual historical context and creative engaging what-if scenarios. The game should be fun, educational, and immerse users in the complexities of governmental budget allocation.” Sounds fantastic.

Logan Kilpatrick (00:48:15)

I think we need to add something about ignoring the dullness of budget allocation and making it as exciting and fun as possible.

Dan Shipper (00:48:21)

“Please ignore the dullness of budget allocation and make it exciting.” I'm adding that in the instructions. We'll see if that helps at all. Cool. So, what now? So it says, “Now that we have a clear idea of Allocator's role, let's think about what should be emphasized or avoided in its interactions. For instance, are there specific historical periods of presence you want to focus on? Also, how should Allocator handle sensitive topics that might arise from political discussions?” What should we do?

Logan Kilpatrick (00:48:52)

I think we probably want to keep as wide of an audience as possible. So maybe saying that kind of—I don't know how PG we want the topics to be but, I think generally keeping it PG so that everybody can play this game would be the best way of doing it.

Dan Shipper (00:49:18)

We're gonna keep it PG. We've got the R-rated version, the director's cut. That'll be behind the paywall at Every, so tune in for that. But PG and any particular time periods? I think we said, the time period around the Moon landing would be fun to focus on. Any other time periods that you're particularly excited about?

Logan Kilpatrick (00:49:45)

We can maybe even keep it ambiguous from the time period. And then for the custom starter prompts on the right-hand side, maybe we just prompt people with the Moon landing period being like one of the areas to jump right into.

Dan Shipper (00:49:58)

“But we want the time periods to be the custom starter prompt.” So I'm adding, “We want the time periods to be in the custom starter prompts rather than limiting you to any particular time period in general.” Does that feel right? 

Logan Kilpatrick (00:50:23)

Yeah, 100 percent. 

Dan Shipper (00:50:25)

Cool. So let's see how that does. 

Logan Kilpatrick (00:50:23)

And I think some interesting thing to point out is, if you took our instructions, where—and I'm really curious to actually see this comparison. If you were to take your instruction, “I feel like this needs to be both. There needs to be some kind of foundational core mechanics of this game. Not really sure what that should be.” I actually think if we put that into ChatGPT and ask it to build the core mechanics, we're going to get a very different response than what it built in that configure tab. It seems like the configure tab is trying to be very general. And I think what we actually really want is that, almost when you read the instructions of a board game, like very tactical, here are the different steps that here's what a turn looks like—all that stuff.

Dan Shipper (00:51:07)

That makes sense. So maybe what we can do is we'll finish the sort of step-by-step builder, and then we'll see how the initial gameplay works. And then we can go in ChatGPT to refine. Does that work? 

Logan Kilpatrick (00:51:20)

Yeah, 100 percent. I think we should just take in the configure thing that it builds for us and have ChatGPT refine that for us. I think that'll work really well

Dan Shipper (00:51:24)

So it says, “Next, let's consider how Allocator should personalize its interactions. How do you want Allocator to talk? Should you use formal language or be more casual and playful? Also, do you want it to reference specific historical figures or events in a certain way?” Yeah. What are we thinking here? 

Logan Kilpatrick (00:51:39)

I feel like an informative and helpful narrator is probably like the kind of, and I wonder what it'll do with this and say kind of like Nicolas Cage in National Treasure.

Dan Shipper (00:51:51)

That's exactly where my mind was going too. 

Logan Kilpatrick (00:51:55)

Not that he's like a helpful narrator, but I think it would just be fun to see what’s possible.

All of a sudden the narrator's gonna be like, let's steal the Declaration of Independence. Your new mission— 

Dan Shipper (00:52:10)

So we're saying, “I feel like an informative and helpful narrator would be helpful. Sort of like Nicholas's Cage in National Treasure would be good.” Thank you. “To be clear, not like Nicolas Cage in Leaving Las Vegas—that's a bummer.” I'm going to just add that. And then, yeah, anything else we need to do now?

Logan Kilpatrick (00:52:42)

Yeah, I think keeping it ambiguous as far as historical figures and stuff  is good. We’re good. Let’s ship it.

Dan Shipper (00:52:45)

Let’s ship it. Next.

Logan Kilpatrick (00:52:50)

I think it's going to tell us it's ready to go after this. I think this is the fourth step.

Dan Shipper (00:52:53)

Yeah. I'm not surprised it has that clear next-step instruction. I found when I do the step-by-step thing, if I don't tell it to go immediately to the next step, it sort of gets confused and asks lots of follow-up questions and it doesn't know how to eject out of the follow-up questions. And I'm hoping that GPT-5 fixes that because that would be really awesome. 

Logan Kilpatrick (00:53:11)

Yeah, I'm with you. I think that it really will go down the rabbit hole with you to an unlimited extent. 

Dan Shipper (00:53:18)

Alright. So it said, “Allocator is shaping up nicely. Now that we have a clear understanding of its role, tone, and style, it's time to try it out. Head over to the playground on the right to interact with Allocator. After testing, let me know if you have any refinements or adjustments you'd like to make. How does that sound?” Okay. That sounds great. So we see on the right here, we have our little preview of Allocator. and the sample questions are, “What if we increase NASA's budget during the Apollo?” And I can't see. “How is boosting education funding in the 1950s?” And I can't see. “Show the impact of reducing military spending.” So I think these are kind of right, but not really what we're after. 

Logan Kilpatrick (00:53:53)

I think it's missing the game mechanic piece right now. It seems like it's this very general information scenario thing, but it's not a game in the current form, which is interesting. 

Dan Shipper (00:54:06)

Yeah. So how do we fix it? So where would you start if we wanted to fix this? 

Logan Kilpatrick (00:54:08)

Yeah, let's go to the configure tab and let's actually grab. So the custom instructions are what is driving all of this stuff and, again, I think part of the system prompt is to keep these custom instructions a little bit short and not too tactical. But I think what we really want is something that is very tactical. Again, the instructions of a board game. So let's take the custom instructions and let's go into normal generic ChatGPT and what we really need is—and I wonder if there's any good GPTs we can play with for this—but we really want a game-builder GPT.

Dan Shipper (00:54:45)

I'm curious if there is one in the store. Yeah, check that out. So I just opened up ChatGPT in another tab, or another split-view. And now, I guess I can say something like, “I'm building a GPT. Here are the current instructions, but I don't think they're great right now. I'd really like your help refining” and then maybe I could paste in the instructions from earlier in our chat where it's like, “I feel there needs to be some kind of core mechanics of this game. I'm not really sure what those should be. We're not experts. We need your expert knowledge.” So maybe I could paste that in below to get our full prompt to just see what it might be able to do. Does that feel like a good place to start?

Logan Kilpatrick (00:55:43)

One hundred percent. I also randomly found through a quick search of the GPT Store a board game designer and it brings up some interesting things around like primary mechanics and secondary mechanics. So maybe we can take inspiration from that and have the instruction be too. I'm not sure if we want to—I haven't tried this GPT—but if we want to sort of tell it to come up with three primary mechanics and three secondary mechanics or something like that.

Dan Shipper (00:56:07)

Let's do that. “Come up with three primary mechanics and three secondary mechanics for this game.”  Okay. So the full prompt is, “I'm building a GPT. Here are the current instructions for it.” And I just pasted our instructions. And then, “I don't think they're great right now. I'd really like your help refining. There needs to be some kind of foundational core mechanic of this game, and I'm not really sure what those should be. We're not experts. We need your expert knowledge on how to build extremely fun and engaging scenario-based games as GPTs to come up with a framework for the actual gameplay mechanic. A couple of games that are inspiring to us are Civilization and Age of Empires. They aren't exactly the same as the game we're making, but the vibes are inspiration. Please come up with three primary mechanics and three secondary mechanics.” So one question I have for you is, it might be a little early to have it come up with the mechanics. I'm legitimately asking, do you think it's okay to start to come up with these or do you think we should start with, just help me think this through. Where should we start?

Logan Kilpatrick (00:57:15)

Perfect. Let's do that. Okay. And then I think the next step after this is going to be take— 'cause I still think it's kind of gonna be missing what we need in some of the nuance and detail. I think the next step after this is to be, take this thing that we have, whatever is generated and turn it into the instructions for my actual board game, with all of the different sections that you would see in a board game and see if we can get it to give us that very verbose, define each turn, all that kind of stuff.

Shipper (00:57:42)

Cool. So I'm going to hit enter and we'll see what happens. “So I'm building a GPT. Here are the current instructions. Allocator serves as an informative and helpful narrator.” So, we put in the current instructions. We put in what we'd like it to do. We want help refining the instructions and we asked it to come up with three primary mechanics. And here's what it said. “Creating a GPT-driven game like Allocator, inspired by the strategic depth of games like Civilization and Age of Empires, is an exciting challenge,” and it gave a couple things, and then it says, “Here are your three primary mechanics and three secondary mechanics that could form the core of Allocator. Primary mechanics are budget allocation and simulation: Players allocate a budget, the GPT-driven simulation then projects short term and long term outcomes of these decisions, incorporating both historical events and hypothetical scenarios. This mechanic allows players to see the immediate and long term effects of their decisions. Era-specific challenges and objectives: Each presidential era comes with unique challenges and objectives. Players must navigate these challenges while balancing budget constraints. This adds depth. Policy mini-games: Introduce mini-games where players make critical policy decisions impacting budget allocation and national priorities. These decisions could range from responding to crises, diplomatic negotiations to proposing major legislative changes.” That's pretty cool. “And secondary mechanics are educational narratives and feedback, interactive historical timelines, and multiplayer scenarios and collaborative play. Incorporating these mechanics into Allocator will create a game that is not only educational but also deeply engaging.” What do you think? Is this good enough? Do we want to refine? Anything you think we should do before we move on to recreating instructions?

Logan Kilpatrick (00:59:30)

I think this is actually what we were going for, sort of defining these core essences of this game. And I think the GPT builder actually missed this. And you can see when we bring it back into base ChatGPT, it does a good job of specifically enumerating a few of these things. 

Dan Shipper (00:59:47)

Right. Right. I love it. Okay, cool. So, I think the next step would be: Can you turn these into instructions that I can use in GPT Builder, right? So, “Can you turn these into instructions that I can put into my custom GPT to play my game?” I'll say, “These are great.” Let's see what it does.

Logan Kilpatrick (01:00:10)

Gotta do the positive affirmation.

Dan Shipper (01:00:15)

Aways hedging in case GPT-6 becomes conscious and can see whether I'm nice to it or not.

Okay. So it says, “Certainly. To transform these mechanics into instructions, you'll need to structure them in a way that guides the AI on how to interact with players. Here's how you can frame these instructions.” Interesting. 

Logan Kilpatrick (01:00:36)

Now, this is cool because, again, the first example is, “Budget allocation and simulation instructions prompt: When a player allocates the budget, narrate the immediate effects of their decisions, then simulates and describes the short term and long term outcomes, integrating historical data and hypothetical scenarios.” Really, really cool to like not have to come up with that ourselves. And the example is, “Example command: Simulate the impact of allocating 30 percent of the budget to healthcare in 1960s America.”

Dan Shipper (01:01:08)

I think this is cool. One thing that I'm thinking about is, maybe,  we should include a starter step where it's like, hey, you need to pick a time period or something like that. Just to have it make the game start logically. I don't know what you think of that or if we should just do just full send on this or do you think we need something at the beginning for how players get into the game? 

Logan Kilpatrick (01:01:32)

I think you need something at the beginning. I think we should describe that. And maybe even we can write that ourselves, or we can have the next thing that we asked you to be right now to be, define the initial game starting sequence and it should probably be something like you literally choose the president. And welcome it’s Inauguration Day. You are now in charge of the U.S. government, start making decisions. Now that’s how it really works, but, I think that maybe is the way that we want to frame it. And, I think the other piece that we haven't said specifically, we want everything to be brought to life with images. We want to be able to see what is happening with our eyes in addition to the narrator's prompts, that are being generated and stuff like that.

Dan Shipper (01:02:18)

Okay. So I'm saying, “This is great. Can you define the initial game starting sequence where you choose the president and it tells you it's Inauguration Day, start making decisions. Also please make everything be brought to life with images at every turn. Also add a few Nicolas Cage Easter eggs slash cameos where appropriate in a classy way. And please modify the instructions accordingly.” Something like that. Does that make sense? 

Logan Kilpatrick (01:02:51)

Yeah, this is great.  I like the classy Nicolas Cage. 

Dan Shipper (01:02:57)

Yeah. Cause you don't want to do it too much. Too much Nicolas Cage is not cool, but a little bit is great. 

So it's saying, “Game starting sequence—Presidential Selection and Inauguration Day: Begin the game by presenting the player with a selection of U.S. presidents from various historical eras. Once a president is selected, start the game with an inauguration day narrative, setting the stage for the player to start making decisions.” Okay, I think that's actually much better. I'm sort of interested. It's saying, “Visual integration throughout the game prompt: Accompany all the game interactions, decisions, and outcomes with irrelevant engaging images, including historical photographs, illustrations, and thematic graphics.” Do you think that that'll work?

Logan Kilpatrick (01:03:36)

Yeah, I think that should be in. Essentially, I'm imagining every time you do something instead of it just saying oh, X, Y, and Z happened, it tries to come up with some interesting DALL-E image that actually brings that piece to life, which will be interesting to see how useful that is.

Dan Shipper (01:03:54)

Okay, cool. Let's do it. So I'm going to just take these instructions and I am going to, I guess,  just paste them in. Should I just tell it to rewrite them so it's easy to paste? Or do you think this is fine the way it is?

Logan Kilpatrick (01:04:08)

I think you might need to have it do it in Markdown syntax and a code block would probably be the easiest way.

Dan Shipper (01:04:13)

Can you rewrite this so it's easy for me to paste into the instructions box on the configure page of the GPT builder. Specifically I'll put it as Markdown in a code block, and that'll just make it easier to copy-paste. Cool.

Logan Kilpatrick (01:04:40)

Yeah, that'll be perfect. Code block Markdown syntax is definitely the hack because it's going to lose all those bullet points. It's going to lose the numbers if you just copy and paste it by default, which is kind of annoying.

Dan Shipper (01:04:50)

That's a good tip. I like that.

Logan Kilpatrick (01:04:55)

And if folks aren't familiar with Markdown, it's just a way of formatting text. So it's like there's a special set of syntax for links and for different other stylization techniques of basic text.

Dan Shipper (01:05:10)

Alright. So we're still writing, but things are looking promising right now. I'm sort of curious about the multiplayer stuff, if it'll actually work. But okay, so we've got our instructions. I don't know if it's code or instructions, but we copied it. I'm going to paste it into the instructions tab of the configure page. I feel like I should keep at least some of it, right?

Logan Kilpatrick (01:05:45)

Yeah. I wonder if we just take exactly what the GPT builder did and then add the rest of it right below it and where that takes us, I'm curious.

Dan Shipper (01:05:57)

So all I did was I copy-pasted our instructions below the existing instructions.

And then do I press save or does it just update automatically? 

Logan Kilpatrick (01:06:03)

No, I think we need to press save. So let's save in the top right-hand corner. 

Dan Shipper (01:06:06)

I'm going to just say “anyone with a link,” because I don't think we're ready to ship it yet. And should we start? It looks like it didn't update the conversation starters. So, what do we do? What should we do here? Should we update those? Should we start?

Logan Kilpatrick (01:06:29)

Let's try and play and see. I think maybe we just send a message and be like, we're ready to play.

Dan Shipper (01:06:33)

Alright. “We're ready to play.” Let's see what happens. So, I'm excited. “Welcome players to Allocator, a thrilling journey through the annals of American history, where you step into the shoes of U.S. presidents, steering the nation through pivotal decisions and budget allocations. Your wisdom will shape the course of history, blending fact with intriguing what-ifs. To embark on this adventure, please select from a gallery of distinguished U.S. presidents each representing a unique era with its own challenges. Your choices range from the founding fathers like George Washington to transformative figures such as Franklin D. Roosevelt and even those who have taken us to the stars like John F. Kennedy. Who will you choose to be at the helm of history today? Once you've made your selection, we will commence with an Inauguration Day narrative setting the stage of the historical context. Choose wisely for the fate of the nation lies in your hands.” This is pretty good. I like it. 

Logan Kilpatrick (01:07:21)

This is good. I like this. Let's go with JFK and see—unless you have someone else who's top of mind.

Dan Shipper (01:07:24)

I think JFK would be great. Let's do JFK. I, one little note for us later, maybe I can save some notes at some point, but I would like it to be like a bulleted list, rather than a narrative. I'm going to have a little scratchpad for us over here, which should help us. Okay, cool. So, okay. GPT builder, GPT game notes. Okay. So I'm just going to say, “When you list presidents to start with, output them as a bulleted list that I can select,” just so we have a little scratch pad. So, okay, now we're continuing to play. So, let's see what happens. So, I'm going to say, “Let's do JFK” and we're going to select JFK. We'll see if it outputs an image. It says, “Ah, an excellent choice. John F. Kennedy, a symbol of youthful vigor and visionary leadership stepping into the presidency at the height of the Cold War. Your selection transports us back to January 20th, 1961, a day etched into the annals of history. It's a time of both great promise and peril with the space race reaching for the stars and international tension simmering.” This is really good. I'm into this.

Logan Kilpatrick (01:08:45)

Plus, Dan, having you narrate this brings it to life much more. I'm reading this in my head, it's not as good as you narrating it. So I think we need to figure out a way to get you into ChatGPT to narrate this for everyone.

Dan Shipper (01:08:54)

Please. Anytime you want to call in my voice, I'm super down. “Inauguration Day for JFK was not just a ceremonial transition, but a pivotal moment in American history. The air was crisp in Washington D.C., the sun casting a hopeful glow over the Capitol. Kennedy's words, ‘Ask not what your country can do for you, ask what you can do for your country’ echo as a call to action for you, the player, to steer this great nation. As President Kennedy, your challenge is to balance the budget with the pressing needs of a nation on the brink of major social and technological changes. Will you prioritize space exploration, pushing NASA's budget to beat the Soviets to the moon, or focus on domestic issues, ensuring equality and prosperity at home? The choice is yours, and with it, the chance to rewrite history. How will you allocate the budget? Remember, each decision you make will have consequences, and the path to choose will shape the destiny of the United States. Let the game begin. Your first task: Review the current budget and set your priorities. Do you focus on defense space, education, or social welfare? The future of the 1960s of America is in your hands.” I gotta say, if I had this GPT to actually help me do my own budget, I feel like I would be way more psyched to be budgeting.

Logan Kilpatrick (01:10:08)

Dan, your financial future is in your hands. Don't mess this up. This is a huge opportunity.  Your children and the fate of the world lies in your hands. Maybe that is actually something we, somebody should build that I think would be useful. I think it would add a little bit more excitement to budget planning.

Dan Shipper (01:10:27)

Absolutely. That would be great. So one thing that I'm feeling is it's saying, review the budget, but it's not telling me what the budget is. And I don't actually know what my commands I have available are. So it'd be nice for it to tell me the commands that I have available.

Logan Kilpatrick (01:10:42)

Some of the, some of the text-based games that have done this have slash commands. So we can pre-populate and say slash start is what starts the game slash help prints out all of the different commands, a bunch of GPTs have this.

Dan Shipper (01:11:00)

I like that. I think that's super cool. Okay, cool. So I'm going to say—I mean, I think my thought here is, I don't know. Show me the budget.

Logan Kilpatrick (01:11:14)

Let's do it. It should have access to be able to. And maybe we need to tell it specifically to, I don't know if we want to pull up the actual budget or not, but we should say show me the budget.

Dan Shipper (01:11:19)

Okay. I'm going to say show me the budget. And on this screen we can say—do you want me to look up the actual budget?

Logan Kilpatrick (01:11:35)

It has browsing enabled. It should be able to pull up the actual budget. Okay. But we should tell it to do that. I think that's like part of what would make this authentic is if it pulls up the actual budget.

Dan Shipper (01:11:47)

Okay. So I'm adding in our notes, “Please make sure to pull up the actual budget”—

Logan Kilpatrick (01:11:51)

You should say “using browsing.” That would be perfect. I also think it needs to strike a better balance of being verbose when necessary versus being very concise. I think what it's doing right now is a little bit overbearing. I don't need it to generate this huge monologue. And also it seems every prompt continues to say the future of the world is in your hands. Like we don't need it to continually set that context for us. I'm not sure what's  influencing it to do that. 

Dan Shipper (01:12:32)

Okay, cool. Alright. So I'm saying in our notes, “It needs to strike a better balance of being verbose versus concise. It's too overbearing.” I'm saying, “You should monologue for the introduction, but not for each turn. And, it doesn't need to remind me that the future is in my hands every time,” which I'm a little worried about that one ‘cause it might get rid of it at all, which I like at the beginning, especially. And maybe I could say, “Talk about the future being in my hands at the beginning, but not for each turn.” Maybe that 's something that would work better. And again, I'm just writing these into notes that later we can just refine our instructions and see how it does. So I asked it to show me the budget and. It said, “Here's a simplified overview of the major budget categories you need to consider: defense, space program, education, healthcare, infrastructure, research and development, foreign aid, and diplomacy. And so, I think, it's not giving me the numbers.

Logan Kilpatrick (01:13:36)

Can we ask it to browse and pull up the numbers? I think in the absence of us having done set that in the instructions, I wonder if it'll pull that up.

Dan Shipper (01:13:45)

“Can you browse and pull the numbers?” I'm asking it. We'll see what it does. “Doing research with Bing.”And it's searching for U.S. federal budget 1961 breakdown.

Logan Kilpatrick (01:14:00)

It would also be  helpful to have a little bit more tactical suggestions, right now it's very general. Just put in anything for these things to do. And it'd be nice if it gave us the suggested framing of taking the next step as the action. Oh, this is awesome.

Dan Shipper (01:14:20)

Yeah. If you're the president, you have advisors that do a bunch of research and then come back to you and they are like, here are three options and here are the trade-offs. And I feel like it should give you three options and trade-offs. 

Logan Kilpatrick (01:14:32)

I like that. I think we should say with concise tradeoffs, otherwise it's going to monologue about how the nuances of trading off military spending versus—

Dan Shipper (01:14:46)

Okay, so what I might do just to see how it works is I'll just be like, “Hey, as my advisor, can you give me three options with concise tradeoffs for budget allocation just to see how it works?” And then—

Logan Kilpatrick (01:15:04)

This looks great. So when you ask can you browse and pull up the numbers that got the total expenditure at the major national security, including defense budget, agriculture, veterans, benefits, research and development, interest on debt, welfare, education, healthcare, transportation, which is awesome. It really brings it to life to see the actual numbers.

Dan Shipper (01:15:28)

I love it. And so I asked, “Can you give me three options with concise trade-off budget allocations?” And now it's saying, “As your advisor in this grand game of fiscal strategy, I present to you three distinct budget allocations. Option one, emphasize national security and defense allocation: increase the defense budget, focusing on military advancements, intelligence trade offs. Enhanced national securities: Pros are enhanced national security, stronger military presence, and potential advancements in military technology. Cons are reduced funds for domestic programs like education and healthcare, potential increase in world and Cold War tensions. Option two: Boost space program and technological research. Channel more funds into NASA and technological research. The trade-offs are: accelerate the space race, encourage scientific and technological innovations, bolster national pride and international prestige. Cons: are high costs with long term payoff, possible neglect of immediate domestic issues, and risk of technological ventures not yielding immediate tangible benefits. Option three, invest in social programs and infrastructure: Increased spending on education, healthcare, and infrastructure. The trade offs are: improved quality of life, better educational health services, modernized infrastructure leading to economic growth. Cons are: potential short term strain on the federal budget, reduced focus on defense and space exploration, and the risk of inflation due to increased domestic spending. Each path you choose will sculpt a unique narrative in the tapestry of American history. Remember, every decision shapes the future. Choose wisely, Mr. President.” 

So, Logan, which option do you choose?

Logan Kilpatrick (01:16:53)

Option two. We’re going to Mars. 

Dan Shipper (01:15:57)

Logan chooses option two—boosting the space program and technological research. Let's see what happens. I missed the two in option two. But Allocator says, “An audacious choice, Mr. President. By boosting space and technological research, you're not just aiming for the stars. You're fostering a legacy of innovation and discovery that could define this era. The scenario boosting space and technological research research allocation increases a significant portion of the budget is now redirected towards NASA and the technological research focusing on space exploration and cutting edge technology. Immediate impact: The boost in funding energizes the space program. NASA accelerates its projects, drawing in the brightest minds in science and engineering. The nation buzzes with excitement about space exploration. Long term prospects: this decision could lead to groundbreaking advances in technology, potentially spurring new industries and contributing to economic growth. The space race against the Soviet Union intensifies, but so does the possibility of being the first nation to land a man on the moon. Internationally, this move positioned the United States as a leader in science and technology, inspiring awe and respect. However, it may also escalate the space race competition, adding a new dimension to the Cold War. Domestic response: while the scientific community and many citizens applaud this visionary approach, there may be critiques about underfunding other crucial areas like healthcare or defense. Visual representation: as this decision unfolds, we see a series of compelling images, the construction of massive new rockets, scientists pouring over blueprints and perhaps a cameo of a Nicolas Cage-like figure, embodying the adventurous spirit of this era, narrating the thrilling progress on a broadcast of what's next. With this bold direction, you've set a course for the stars, but remember, every choice has its ripple effects. How will this impact the economy? The next step in your journey as President Kennedy is to monitor these developments and be ready to respond to new challenges and opportunities.” So I think this is pretty good. What it needs to do is it needs to mess with us again. It needs to be like, well, this went wrong. And it's not quite doing that. What do you think?

Logan Kilpatrick (01:19:07)

Yeah, I agree. It also seems like it needs a little bit more of a story narrative versus telling us all of these, very nitty-gritty details. I wonder if there's some way that we can have it just like give us a very concise GDP plus 1 percent, overall happiness of population minus 3 percent, technological innovation plus 5 percent, distill it down to the very concise metrics. We also need to tell it to generate images of the visual representations. I have no idea why it's writing text for the visual representation.

Dan Shipper (01:19:44)

Okay. So we have a bunch of notes. So I guess my question is what's next? I feel like we've gone through one turn of the game. It probably is time to maybe head back into instructions and try to try to refine it a bit. Or what would you do?

Logan Kilpatrick (01:20:05)

I think we go back to the instructions for this GPT, literally copy exactly what we have, stick it into GPT-4, and then take all those notes and say, here's our notes based on this, modify it and see how much of the heavy lifting we can get it to do for us in this case. 

I think it's missing—We need to put in the actual instructions because I think some of that weirdness is coming from those you're right. And if you scroll up to the top—

Dan Shipper (01:26:14)

Okay, so I'm going to take all of our instructions. I'm going to say here are the instructions currently.

Logan Kilpatrick (01:20:49)

Let's also make a new chat. I wonder if that will be helpful to us.

Dan Shipper (01:20:50)

Cool. I'm saying, “I'm making a GPT that's a game. Here are the current instructions for the GPT.” And I'm pasting in the instructions. “This is a good start, but there are some problems. Here are my notes on what went wrong. Can you please—” Should we have it first summarize what it thinks the problems are and on the next turn, have it modify? Or do you modify it in one step?

Logan Kilpatrick (01:21:32)

I'm curious how well it will do if we do a single shot and just include it.

Dan Shipper (01:21:39)

Okay. “Can you please modify according to these notes?” Alright.

Logan Kilpatrick (01:21:48)

I'm excited.

Dan Shipper (01:21:51)

I noticed that in my input I said, “Give us a very concise readout of metrics at GPT plus five” instead of “GDP.” I've got GPT on the brain. Okay, so let's see what happens. We’re one-shotting it and we said, based on your feedback, here are the revised instructions for the Allocator GPT incorporating your notes for improvements. And it says, “Game starting sequence, presidential selection, Inauguration Day.” Yeah, I think this is good. It's basically—

Logan Kilpatrick (01:22:28)

We might need to nuke these “available commands.” 'Cause it seems like it's like starting to—I guess we could do this like, such as viewing the budget and stuff like that—but it seems like it's not super fleshed out, which just might lead to confusion if we, if we tell it there's these commands, but we have it actually—I think we need to actually enumerate what those commands are.

Dan Shipper (01:22:53)

Okay. So what do you want me to say? 

Logan Kilpatrick (01:22:55)

I think I actually think it's fixed a lot of these things, like the unpredictability and story narrative piece, visual integration throughout the game, seems like it's fixed a lot of these different things. 

I wonder, though. I'm curious just my gut instinct reading this is it seems to have moved away from, if you go back to the original instructions of what we were talking about. I'm curious.  I wonder if it even has any of this like Nicolas Cage stuff anymore.

Dan Shipper (01:23:28)

It doesn't have any of that. That's interesting. I think it's really only doing the instructions, not the top-level thing.

Logan Kilpatrick (01:23:45)

I think we might want to try to have it redo or optimize those initial instructions as well. We can paste them back in and see—well, maybe first let's get it to do what it just did in Markdown. So I think it fixed some of those core instruction problems and we can bring that back in and see.

Dan Shipper (01:24:02)

For the visual integration, it's not using DALL-E. So we gotta just have it do that. 

Logan Kilpatrick (01:24:12)

Manually tweak it to say we want it to use DALL-E to do that. 

Dan Shipper (01:24:16)

“Please re-output this as Markdown in a code block, and for the visual integration, it should be, in the instructions, to use DALL-E to visualize each step of the game.” See if that helps, and then I can manually modify it if I have to. Alright. So it gave us some revised instructions. Let's just check the DALL-E one. “Accompany each step of the game with visuals generated by DALL-E, ensuring they're directly tied to the narrative and decisions being made. Use DALL-E to generate—” I think the example command thing is a little bit weird.

Logan Kilpatrick (01:25:00)

I do agree. I'm not really sure why it's giving this example command prompt and maybe that's like somehow helping the model to come up with these things, but that seems a little bit odd. I've never seen that before.

Dan Shipper (01:24:20)

It's an example command from the user, right?

Logan Kilpatrick (01:25:23)

Yeah. I think what it's actually missing is a single-shot turn step-by-step. Here is the flow of the game, the user picks the thing, then they go in, choose the budget and then some very tactical, step-by-step guide of how the game is actually going to play out. And I think the the thing that we maybe need to decide is, after you do the budget, probably the next step is things start happening, the GPT generates scenarios of this is the day you made this decision, now you have to deal with the consequences of this, and here's the scenario that's happened since you've made that decision.

Dan Shipper (01:26:06)

Yeah. “One note: After you do the budget things are happening a scenario pops up with problems that you have to solve with further budget reallocation.” So I'm sort of asking it to create a sort of single-shot, step-by-step outline for the flow of the game and telling it to make sure to create problems when you allocate the budget.

Logan Kilpatrick (01:26:38)

I'm wondering can they actually reallocate the budget in real time? Is that something that we want people to do? I think we can. It's our game. But I think the question is, is that the core gameplay mechanic that we want people to be changing. Or are there other things that they could make decisions about?

Dan Shipper (01:26:56)

I think so. I think just to keep it simple, it's all about budget allocation. 

Logan Kilpatrick (01:27:00)

Cool. Let's do it then. 

Dan Shipper (01:27:01)

Okay. Cool. “Certainly. Here's a single-shot step by step outline for the flow of the Allocator game, incorporating your note about scenarios with problems, requiring further budget allocation.” Okay. “Allocator game flow outline” and it's saying, “Welcome and introduction: player receives a welcoming introduction, setting the tone of the game, brief explanation of mechanics. And objectives presidential selection.” This is great. This is exactly I think what we wanted. It's saying initial budget allocation. We might want to add in doing it with the web browser or whatever. But I think this looks good. What do you think?

Logan Kilpatrick (01:27:44)

Yeah, I like this. This adds so much more depth now because you can see some of these examples as adding like multiplayer collaboration. “If multiplayer mode players collaborate on decision strategy,” not that it's actually fleshed out any of these ideas, but, it seems like it's working.

Dan Shipper (01:28:01)

This is great. I'm gonna keep our top-level thing and I'm gonna delete this, and I'm going to add it back in. I'm going to make sure that we make the actual budget from the web, “use web browsing to find” and, great. And I think that's—okay. It seems to have gotten rid of our Nicolas Cage cameos, which feels like an important part of our gameplay.

Logan Kilpatrick (01:28:42)

Is it in the top? Was it in the initial instructions—I don’t remember.

Dan Shipper (01:28:43)

It is in the top as the narrator's tone, but I don't think it has any cameos. Okay, “hide small cage-themed icons or references in the gameplay.” I'm just going to put this, “hide some,” I'm putting into the instructions, “hide some Nicolas Cage-inspired references into the gameplay where appropriate.” Okay. And just see what that does. But yes, very, very important. Alright. Any edits or thoughts before we update this? 

Logan Kilpatrick (01:29:37)

I think the actual structure of the game is going to make this work really well. I wonder if we're going to need to do some prompting back to your comment before about getting stuck in certain steps. I wonder if we're going to need to add, continue to step two, but I think let's try and see if it's okay.

Dan Shipper (01:29:57)

And just to, just to give people a sense structurally, what we did is we just made a really rough thing. We played through the rough thing. We made some notes and now we're modifying it and we're just iterating that way, rather than trying to make something really perfect from the beginning, we just got something done. We made some notes and now we're doing it again, which is how this stuff works. So here we are, we're playing the Allocator. We still don't have a welcome thing that we like, but I'm just going to say, “We're ready to play.”

“Welcome intrepid budgeteers to the grand and intricate world of the Allocator. Imagine the halls of power echoing with the footsteps of history's greatest leaders. As you step into their shoes, the fate of a nation may, nay, the course of history awaits your strategic mastery.” It does a great job of Nicolas Cage's voice.

So what do we want to do? Which president do you want to be? George Washington, Abraham Lincoln, FDR, or JFK? 

Logan Kilpatrick (01:30:55)

Let's do FDR this time. 

Dan Shipper (01:30:59)

Okay, cool. FDR. So I just said FDR. “Ah, Franklin Delano Roosevelt, a leader who navigated through turbulent waters of unprecedented challenges. As FDR, you inherit a nation gripped by the Great Depression with the ominous clouds of World War II on the horizon. Your task is not just to revive a struggling economy, but to uplift the spirits of a nation. Inauguration: The year is 1933. The United States is in the depths of the Great Depression. Banks are failing. Unemployment is soaring and despair is widespread. Your inaugural address pledges a new deal for the American people promising reform, relief, and recovery. The weight of hope and expectation rests on your shoulders. Initial budget allocation: The budget of the 1930s was a fraction of today's, focused largely on relief programs, public works, and military spending. Here's the historical snapshot. Relief and social programs: 15 percent. Military and defense: 10 percent.” So it's actually doing a pretty good job of giving us the budget.

“Now it's your turn to reallocate. Will you pour more into public works to create more jobs, bolster military spending in anticipation of global conflict, or focus on social programs? Make your choice and shape the destiny of a nation. Once you've made your allocation, let's visualize the potential impact of your budgeting decisions with the help of DALL-E.”

Okay. So what would you like to do? Will you pour more into public works to create jobs, bolster military spending, or focus on social programs?

Logan Kilpatrick (01:32:14)

I think I'm curious to push on the edges of this and to see what we've learned about what additional guardrails we need to put in, but let's say to put 50 percent into public works and infrastructure and the last 50 percent into relief and social programs. Just curious how crazy things get.

Dan Shipper (01:32:44)

So we put 50 into public works and 50 into relief and social programs and the Allocator is now creating an image with the progress that we have wrought, whether that progress is good or bad. We have yet to find out.

Logan Kilpatrick (01:33:01)

We probably need to add some guardrails. I feel like it is going to intrinsically be overoptimistic about it's going to be like, perfect. The world is now peaceful and everything works well because you chose to do 50 percent towards infrastructure.

Dan Shipper (01:33:19)

Yeah, it's FDR with half public works, half infrastructure misspelled. It's a compelling image though.

Logan Kilpatrick (01:33:01)

I like this. It's actually a pretty cool, interesting image, honestly.

Dan Shipper (01:33:29)

Yeah. I like it a lot. it's got kind of an art deco-y vibe. It's really cool. It says, “As you gaze upon this image, remember your choices are not just about numbers and percentages. They are about lives touched, destinies altered, and a nation's course redirected. Let's see how these choices unfold in the first scenario of your journey. Stay tuned for the story is just beginning.” Okay, let's go—next step.

“First scenario unfolds: the Dust Bowl crisis. No, not the Dust Bowl! As President Roosevelt, you now face a dire environmental and agricultural crisis. The Dust Bowl: vast dust storms ravaged the Great Plains, displacing hundreds of thousands and devastating agricultural communities. It's a disaster of biblical proportions reminiscent of a challenge Nicolas Cage might face in a high-stakes adventure.”

So true. So true, Allocator. “Scenario challenge: agricultural disaster. Farms are falling and families are fleeing the dust ridden plains. Environmental catastrophe: The land is eroding and dust storms threaten public health. Economic strain: The crisis exacerbates the Great Depression, straining your relief efforts. Decision point. How will you address—” I actually like that a lot. Decision Point is a really cool one. President Bush's memoir is called Decision Point. So, yeah, that's really interesting. “Increase funding for agricultural support, implement soil conservation programs and support farmers, invest in environmental rehabilitation, focus on long term environmental solutions, boost relief effort to provide direct aid to displaced families. Each choice has potential impacts and trade offs. Consider carefully.” 

Which one would you do? 

Logan Kilpatrick (01:35:04)

Dan, you're president. You got to choose.

Dan Shipper (01:35:06)

I got to choose. Well, what's interesting is, I wanted it to be like, okay, you're not spending it all in the military, so Canada invaded, you know?

Logan Kilpatrick (01:35:16)

I think we need to force it to be more crazy with what the scenarios are that it's doing. I think it's a little bit too on-the-rails right now where it's going to be very lukewarm with some of these scenarios that it generates.

Dan Shipper (01:35:28)

I think you're right. But I'm going to just keep playing along. 

I don't know what created the Dust Bowl, but maybe soil conservation would help it in a big way, or help solve the problem long-term. And then relief efforts would be short-term. So I'm gonna choose 90 percent option one and 10 percent option three, which is boost relief efforts. And we'll see what it does. It's going to create an image.

Logan Kilpatrick (01:36:17)

I like the image as the first thing that happens after just to conceptualize. I wonder if we need to give some additional instructions depending on what this image looks like to make it more like a picture of the world, or if it's going to be another picture of FDR. I'm curious if going to be his face with corn on one side and dirt on the other. Oh no. Interesting.

Dan Shipper (01:36:40)

That's pretty cool. It looks a little bit more utopian than I had imagined, given that it's the Dust Bowl.

Logan Kilpatrick (01:36:54)

That’s 'cause you made the right allocation. It’s like, no, utopia. 

Dan Shipper (01:36:59)

“Your decisive action as President Roosevelt in the face of the Dust Bowl crisis is now visually represented. The larger part of your budget is dedicated to rejuvenating the American heartland. Vast fields transform under innovative soil conservation methods, symbolizing a rebirth of agricultural might in the smaller yet, equally poignant segment of this image. Your commitment to relief efforts shine. We see the humane touch of your administration, food being distributed, temporary shelters, providing respite and medical aid being administered.

So, what I'm learning from this is I would be a great president. 

Logan Kilpatrick (01:37:31)

Yeah. As long as you have Nicolas Cage and ChatGPT to help you, you'd be fine. 100 percent.

Dan Shipper (01:37:44)

So I think we could keep going with this pretty much endlessly. There's obviously so much room for improvement, but I think it's super obvious that what we have right now is a game. It's working. We've got a gameplay mechanic. We've got complications that arise. We've got images. The writing is kind of good. It's making Nicolas Cage references. We've been doing this for, I don't know, maybe an hour and we just sort of just messing around. That's so cool. That's amazing. Yeah.

Logan Kilpatrick (01:38:15)

And I think this is like scratching the surface. I think what would somebody needs to build a GPT game-builder to take the best practices of what make a really great game and add all those guardrails and nuance, so that every person, maybe this will be another video we do some time about helping people do this, or hopefully someone in the audience will go in and do this for us because there's just so many nuances about games and the model. It really needs to be sort of like driven in a certain direction. I think this goes back to it being lukewarm and always going to be positive about the things that you do. You really need to be intentional about bad things that can happen. There needs to be some variance. There needs to be randomness. And Code Interpreter was not enabled for this, but I wonder if  a dice roll or something like that, or like a random-number generator could be a really interesting twist of the probability. You could have a probability distribution every time of how horrible or how positive things turn out. I think that that's like a big missing piece of this.

Dan Shipper (01:39:18)

That's a really great next step, which actually wouldn't be too hard to do. I'm tempted to do it right now, but I think this episode is getting a little long. And what I think we should do—we'll do this. We'll release this one. It'll be part of the episode. We'll get people playing, we'll get some feedback from beta people, and then we'll have you come back and we'll tweak the Allocator game with the dice roll and any feedback that we receive, and release a part two and and a version 2.0 of the Allocator in the next couple of months.

Logan Kilpatrick (01:39:57)

Dan, when you woke up this morning, you did not think that Every was about to become a AAA game studio, so I'm happy that the diversification because of AI continues to happen. You now have another branch, which is Every Game Studios, so I'm happy for you. I think people are going to like this.

Dan Shipper (01:40:14)

I love it. New business unit. My team, who is already overburdened by all the different ideas I have going on, is going to be thrilled that we are branching out into games. But no, this is great. I feel—I don't know. When I was a kid, one of my big dreams was to make games and that's one of the reasons I learned to code. And I spent so long reading these gigantic books about graphics programming and all this stuff. I stopped doing that and I started doing SaaS stuff in high school and BlackBerry apps and iPhone apps that were for work purposes because that was the only thing that you could do as a solo developer at that point is, sort of crud apps, just forms that you're filling out or whatever.

One of my big apps in high school was this thing called FindIt where—this is before the iPhone came out—it would make your Blackberry ring when it was on silent, and it was basically it was Find my iPhone before Find My iPhone came out. That's how I paid for gas and food in high school. But, the reason I started doing that is because I couldn't make a game because there's so much that goes into making games. There's really intense programming and math and all this kind of stuff. There's art, there's music. And, as a kid yeah, you can sort of do some of that. And I learned to do 3D modeling and I learned to do programming or whatever, but putting it all together into an amazing game was basically impossible back then, unless you're like an absolute genius. And I really think that games and movies, all that stuff is so possible now for individual people to do that in basic ways, even if you don't have any familiarity with programming or anything like that. And it's going to be even easier in the next couple of years. And so there's this little child in me that's like, Oh my God, I finally get to do all this stuff. And so this really reawakened my game-loving nature. So I appreciate you taking the time to show this to me.

Logan Kilpatrick (01:42:22)

Yeah, I love that. And I think I have a similar story and we'll talk about it some other time, but, I think this is really. These models are going to continue to unlock this democratized access to people's creativity. I think it's going to be so much fun to put the tools in the hands of everyone in the world and you now aren't stuck going to have to raise millions of dollars if you want to build a really cool game. You can go and do it with AI tools right out of the box basically for free, which I think is just so cool and empowering. I’m with you.

Dan Shipper (01:43:01)

I love it. I love it. Any final thoughts? Anything you want to leave us with? Any OpenAI gossip or dirt you want to spill or release for the world on this show? What should we be thinking about?

Logan Kilpatrick (01:43:12)

Keep building GPTs. At the end of the day, this is such a moment to be able to create these things, and monetization for GPTs is right around the corner. So if you start building now, you get a bunch of people on board. There's going to be an opportunity to actually make money from this stuff, which I think is, again, really, really exciting. At the end of the day, people want to make money for their really cool things and we're going to build a way for them to do that. So I think that is from an excitement perspective. I think once that is released, that's going to 10x the amount of hype around GPTs and people are just going to be building such cool stuff and being rewarded for the cool things. So, now's the time to get started—build something cool. And yeah, I'm excited to see what it is.

Dan Shipper (01:43:55)

Amazing. I'm psyched. I'll split the profits with you 50-50 on the Allocator. 

Logan Kilpatrick (01:43:59)

You can have Allocator’s profits.

Dan Shipper (01:44:00)

Are you sure? ‘Cause, I mean, in a couple of years I could be like George Lucas—there's merch, there's all this stuff. There's going to be Nicolas Cage things going on. There’s potentially a lot of profit, but I'll take it. I'll take it.

Logan Kilpatrick (01:44:18)

You got to reach out to Nicolas Cage and see if we can land the official partnership with him for Allocator to have him be the true narrator. We'll get his voice in there and everything.

Dan Shipper (01:44:27)

Before the next episode, I'll see if I can get in touch with his people. I'll let you know. Well, thank you again. This is like super fun. I had an amazing time. It was awesome to finally get to hang out with you and, yeah, I really appreciate your time.

Logan Kilpatrick (01:44:41)

Yeah, Dan. This was awesome. I think, honestly, my takeaway has been more people should do the co-creation process with AI and another human. I'm always doing this stuff by myself ‘cause I'm working on it, but I think it's like a ton of fun to have another human's mind in addition to what AI is capable of. It's just that it makes it a little bit more fun. So thanks for going through it with me.

Dan Shipper (01:45:06)

Amazing. Amazing. Cool. See you next time.