John: Welcome to The Flux, where we hear stories from people who've asked "what if" questions to better understand the world and talk about how data can help tell stories that impact decisions and create an intentional impact on the future. This is your host, John Cordier, CEO at Epistemix. In a world where the flux capacitor from Back to the Future doesn’t yet exist, people have to make difficult decisions without always knowing how the future will play out. Our guests are people who've taken risks, made decisions when uncertainty was high, and who have assisted decision-makers by using data and models. We hope you can turn lessons from our podcast into foresight so you or your organization can make better decisions and create an intentional impact for others.

Welcome to the next episode of The Flux. Today's guest is Sam Arbesman, scientist-in-residence at Lux Capital and Research Fellow at the Long Now Foundation. I met Sam about four years ago, and since then, I tell people Sam has the coolest job on the planet that I’ve uncovered so far. He basically gets to operate at the intersection of technology, complexity, science, venture capital, and through his writing, helps nudge the world forward in a positive direction. I’ve always been a huge fan and always eager to learn more about what Sam’s working on. Sam, glad to have you on today.

Sam: Thank you very much. It’s good to be talking with you, and I appreciate the very kind words.

John: Absolutely. With all the different pieces of your background, it was tough to find a lead-in question, but between the books you’ve written, the lists you’ve put together, and your other writing, one of the things I like to start with for new guests is: what got you into complex systems or computational social science in the first place? What sparked that interest for you?

Sam: That’s a great question. I don't think there was any single spark or inspiration, but there were a number of different through-lines and experiences that got me thinking about these things.

One was playing the computer game SimCity. My first experience was with the black-and-white Macintosh version, and then I eventually got SimCity 2000, the isometric 3D version, which I still think is the most beautiful of them all. As a kid, I didn’t know anything about city design, but being able to build a little society that actually responded that had feedback loops and emergent behavior was incredibly powerful and exciting to me. Whether or not it was realistic is another question, but the experience of laying things down and seeing how the system would react was amazing.

I remember the SimCity manual had essays and a bibliography about city design, and I convinced my parents to take me to the university library to learn more. I was blown away.

Another touchstone, very different, was reading the Foundation trilogy by Isaac Asimov. I’ve been a huge consumer of science fiction from very early on, helped by my grandfather who read science fiction since the genre's modern beginnings. I still have his original Foundation trilogy copy it’s one thick edition with all three books.

The idea of psychohistory the notion that while you can't predict individual behavior, large groups exhibit regularities that can be modeled struck me profoundly. It made me wonder: are there real versions of this? Computational social science isn’t exactly psychohistory, but it does aim to place bounds on our uncertainty, and you’ll find that a lot of computational social scientists cite Foundation as an early inspiration.

John: Absolutely. One of my co-founders, Don, actually cites the Foundation trilogy as his inspiration too.

Sam: That’s fantastic. And honestly, when I was reading Foundation in the early days of the internet, I searched for anything about psychohistory and found very little just some references to psychoanalysis. I was disappointed. But eventually I found essays, like one in Analog Science Fiction and Fact, about whether psychohistory could be real.

Between SimCity, Foundation, and later learning about complexity science, the Santa Fe Institute, agent-based modeling, and emergence, I realized there was a real intellectual framework behind these ideas. They weren’t just fun games or fiction; there was real substance there. That realization got me very excited.

John: At one point SimCity had something like 40 million active players. When you put your innovation and venture capital hats on, do you see a pattern where science fiction or video games spark real-world technological development?

Sam: In some cases, yes not always companies, but certainly products. Many science fiction writers would say they aren’t trying to predict the future; they're running thought experiments about the present. Still, there are instances where sci-fi inspired real-world inventions.

At Arizona State University, there’s a Center for Science and the Imagination that tries to connect scientists, engineers, and science fiction writers to boost creativity. The first flip phone, for example, was inspired by the communicator from Star Trek. If you look at Star Trek: The Next Generation, you see characters holding what look eerily like iPads.

Spielberg’s Minority Report is another case. Before making the film, he gathered futurists and technologists to extrapolate future trends, and some of the movie’s interfaces and technologies have since materialized.

At Lux, we talk a lot about bridging science fiction and science fact finding people who are thinking that way and helping make their visions real. Sci-fi is great not just for imagining gadgets but for thinking about the societal, ethical, and regulatory impacts of technology once it's widespread and messy, not just shiny and new.

John: We got introduced through one of our favorite near-future sci-fi novelists, Eliot Peper. He also introduced me to the Long Now Foundation. Who else do you think does a great job at creating positive North Stars, not just dystopias?

Sam: I would hope everyone aims for positive visions. The Long Now Foundation is a major influence in that regard encouraging long-term thinking beyond quarterly cycles and into decades, centuries, or millennia.

There are pockets of people everywhere working on these ideas. The key isn’t necessarily predicting exactly what will happen, but expanding our imagination. As William Gibson said, "The future’s already here, it’s just not evenly distributed." Finding those hints of the future and scaling them up is important.

Venture capital plays a role here too. At Lux, we see ourselves as helping identify those hints of a positive future and making them more real.

John: Has venture capital shifted over your nine years at Lux to focus more on long-term challenges, rather than short-term returns?

Sam: Definitely. There’s been a big shift. Deep tech and frontier tech anything with a hard science basis used to be less popular. Now a lot more firms are investing in these areas, which is great.

At the same time, there’s been a boom in new organizational forms to support cutting-edge science outside traditional academia or corporate labs. There’s more creativity now in how science and innovation are funded and structured, which is very exciting.

John: One of our favorite questions: if you had a flux capacitor and could rewind to a “what the flux” moment a time when a major decision was made without enough information what moment would you go back to, and how would you rethink it?

Sam: That's a great question. Nothing super specific jumps to mind immediately, but more broadly, I think we often mistake our models for reality. In complex systems, models are powerful but can never fully capture reality. Maintaining humility is crucial.

For example, during the early stages of COVID, many people's intuitions failed badly about nonlinear phenomena like exponential growth. Simulations and models could have helped anchor thinking much better early on.

John: How do you see the use of counterfactuals or simulations evolving? Are there fields where you think they should be used more?

Sam: I love counterfactuals, both in fiction and in real-world scenario planning. One challenge is that we often conflate outcome quality with process quality a good decision process can still lead to a bad outcome due to luck, and vice versa.

Using simulations and counterfactuals to expand our imagination of what’s possible and then stress-testing decisions against a wide range of plausible futures is critical. I’d love to see this more widely used in policymaking, medicine, aerospace really any field where major decisions are made under uncertainty.

Sam: I think any time you're tasked with making large-scale decisions under uncertainty, you should be using tools that help you narrow down your uncertainty simulations, counterfactuals, scenario planning. If you don't, you're just making stuff up, and that’s not great.

John: One of the people involved with Epistemix talks about how, once a decision is made, it becomes about execution and building conviction behind that decision. From your experience, are there any tools or approaches that help bridge that gap from making a decision to actually building momentum behind it?

Sam: Unfortunately, I’m not sure I have a great answer for that. I think it’s a really important question, though, because even a good decision needs strong execution and narrative to succeed.

One thing I would add, related to your earlier "what the flux" question, is the pandemic showed how our intuitions often fail us with nonlinear systems. Simulations and modeling can help a lot there helping people grasp exponential growth and tipping points more intuitively.

John: Within complexity science or venture capital focused on deep technologies, are there any commonly held beliefs that you think are wrong where your perspective or Lux’s or Long Now’s perspective is different from the mainstream?

Sam: Definitely. I think we still underestimate just how complex and nonlinear our technologies have become. We're enmeshed in systems that are incredibly sophisticated infrastructure, utilities, data centers systems that have almost organic levels of complexity.

This is something I wrote about in Overcomplicated. We can’t treat these engineered systems as if they're fully understandable anymore. Even the people who build them often can't fully explain them. It's more like biology now messy, interconnected, and emergent.

And yet, there's still a tendency to assume that because humans engineered something, it must be fully knowable. That's just not true anymore. I think we need to apply complexity science and biological thinking much more to how we understand modern technology.

John: I remember reading some of your articles about those topics. Might be a good segue can you give us a hint about your third book? Is it about AI, emergence, knowledge over time?

Sam: Yes. My first book, The Half-Life of Facts, was about how knowledge changes. The second, Overcomplicated, was about incomprehensible technology. The third is going to be about the wonders and weirdness of computing and computation very broadly construed.

The idea is that if you take computation seriously, it doesn’t just inform engineering it changes how you think about philosophy, biology, language, even reality itself. Computing is almost like a new liberal art.

AI will play a role in the book, of course, but so will things like ancient storytelling traditions, mythology, and other ways humans have grappled with information and knowledge. It's really about how computation connects across everything and how that shapes how we think about complex systems, emergence, and reality itself.

John: Earlier we talked about how it would be great if more people could think in a complexity-oriented way. If someone wanted to start getting into this kind of thinking middle school, college, even later where would you recommend they start?

Sam: One wonderful thing about complexity science is that even though the underlying math can be sophisticated, the basic intuitions are very accessible.

There are tons of computer models and simulations online simple ones about forest fires, flocking birds, basic emergent behavior that you can play with. Just seeing how simple rules create unexpected patterns is incredibly powerful.

I actually think we should be teaching these ideas much earlier and more broadly. You don’t need to become a complexity scientist, but everyone would benefit from having an intuition about nonlinearity, emergence, and interconnected systems. They're as fundamental as understanding something like Newton's laws of motion.

John: Computer science education has been pushed earlier into schools do you think complexity education could follow a similar path?

Sam: Absolutely.

Think about calculus when Newton and Leibniz developed it, it was cutting-edge math. Now it's taught in high school. We should do the same with complexity science. We’re already living in a complex world, and even basic models can give people handles to understand it better.

I also think we need better tools. One idea I love is creating a “HyperCard for simulation” a low-threshold, high-ceiling tool that lets people easily wire up simple simulations and gradually build more sophisticated ones. Something intuitive enough for beginners, but powerful enough to scale with their skills.

There are some early versions like NetLogo but I'd love to see even more creativity here. Building intuition about complex systems should be as normal as learning algebra.

John: That’s a great point. Final question: over the past year, what are some developments that have made you most optimistic about the future of simulation, computational social science, or complexity research?

Sam: The sheer amount of computing power now available to individuals is incredibly exciting.

Things that would've been impossible or prohibitively expensive even five or ten years ago are now within reach. We still need better interfaces and better software, but the foundational platform cheap, massive computing is here.

That opens the door for all kinds of innovation and exploration in modeling, simulation, complexity science, and beyond. I’m really excited to see what people build with it.

John: Sam, thank you so much for taking the time to be here. There’s a ton of insight here that I think will be helpful to a lot of people.

Sam: Thank you, John. I really enjoyed the conversation.