Welcome back to Scaling Theory. Today, we are taking on a surprising but deeply relevant topic: religion. We are not entering the realm of theology, but rather looking at religion the way an economist might look at a multinational corporation or a digital platform. Think of it this way: in the U.S. alone, faith-based organizations generate more annual revenue than Apple and Microsoft combined. So when we ask how religions scale, we are really asking how some of the world’s most enduring (and powerful) institutions grow, adapt, and persist.Our guest is Paul Seabright, Professor of Economics at the Toulouse School of Economics and author of The Divine Economy: How Religions Compete for Wealth, Power, and People.Paul and I talk about how religions scale, why rituals, doctrines, and compelling narratives matter for growth. We explore how religions act as multi-sided platforms, how they build robust networks that resist churn, and how technologies like the printing press and social media can reshape their reach. Toward the end, we explore whether new movements in the Silicon Valley function like new religions, and what their chances of success might be in today’s competitive market for belief. I hope you enjoy our discussion.You can follow me on X (@ProfSchrepel) and BlueSky (@ProfSchrepel) to receive regular updates.
--------
1:15:39
#18 – James Evans: Science in the Age of AI
Today’s episode is different from all the previous ones, as for the first time on Scaling Theory, we focus on research methodology, exploring how AI is reshaping the very process of doing research and what that shift means for science and society at large.I sat down with James Evans, Professor of Sociology, Computational and Data Science at the University of Chicago, External Professor at the Santa Fe Institute, and Faculty Member at the Complexity Science Hub in Vienna, to explore how AI is transforming the way we simulate, scale, and understand human behavior, and what that shift means for science and society.We dive into his pioneering work on using large language models to simulate individuals, societies, and entire social systems. James and I explore the strengths and limits of AI agents for both the social and hard sciences before reflecting on the future of social science itself. We talk about research centers entirely run by AI and conferences conducted by AI agents, without any human involvement. We also discuss the role of small research teams in disruptive innovation, and how to cultivate proximity and serendipity in a research world where we increasingly cooperate with machines.You can follow me on X (@ProfSchrepel) and BlueSky (@ProfSchrepel) to receive regular updates.References:- Simulating Subjects: The Promise and Peril of AI Stand-ins for Social Agents and Interactions (2025) https://osf.io/preprints/socarxiv/vp3j2_v3- LLM Social Simulations Are a Promising Research Method (2025) https://arxiv.org/pdf/2504.02234- Large teams develop and small teams disrupt science and technology (2019) https://www.nature.com/articles/s41586-019-0941-9?wpisrc=- AI Expands Scientists' Impact but Contracts Science's Focus (2024) https://arxiv.org/abs/2412.07727- The Paradox of Collective Certainty in Science (2024) https://arxiv.org/html/2406.05809v1?utm_source=chatgpt.com- Being Together in Place as a Catalyst for Scientific Advance (Research Policy, 2023) https://www.sciencedirect.com/science/article/pii/S0048733323001956
--------
40:56
#17 – Eric von Hippel: Freeing Innovation
My guest today is Eric von Hippel, Professor of Technological Innovation at the MIT Sloan School of Management. Eric is the author of numerous academic articles and books, including Free Innovation, Democratizing Innovation, and The Sources of Innovation, all published by MIT Press and available for free. Eric has accumulated over 90,000 citations on Google Scholar and has received many awards, including the Schumpeter School Prize (2017)—a particularly interesting recognition given his work on non-Schumpeterian innovation.In our conversation, Eric and I explore the role of free innovation in today’s economy. Eric highlights some of his favorite examples of free innovation and discusses how, despite being developed at personal cost, it is scaling at an impressive rate. We explore the mechanisms that best enable this scaling—whether through recognition, institutional support, IP protections, or alternative incentives. By the end of this talk, you will understand what free innovation is, how it develops, and how it interacts with producer innovation.You can follow me on X (@ProfSchrepel) and BlueSky (@ProfSchrepel) to receive regular updates.References:Sources of Innovation (1988) https://web.mit.edu/evhippel/www-old/books/sources/SofI.pdfDemocratizing Innovation (2005) https://direct.mit.edu/books/book-pdf/2425023/book_9780262285636.pdfFree Innovation (2016) https://library.oapen.org/bitstream/handle/20.500.12657/26044/1004041.pdf
--------
36:21
#16 – David Krakauer: Scaling Intelligence
David Krakauer is an American evolutionary biologist. He is the President and William H. Miller Professor of Complex Systems at the Santa Fe Institute. As you will hear in today’s episode, David's research centers around a series of fundamental questions, such as: How did life and intelligence evolve in the universe? How do ideas evolve and how do they encode natural and cultural life?In this conversation, David and I explore the evolving landscape of complexity science. We discuss its foundational theories, emerging patterns, and intersections with AI and machine learning. We delve into the paradigm shift complexity science represents, its most significant contributions across disciplines, and how computational advances are reshaping its trajectory. We also talk about AI’s potential to scale towards AGI through a complexity lens, the limits imposed by evolutionary principles, and what this means for artificial systems. Finally, as President of the Santa Fe Institute, David discusses SFI’s unique interdisciplinary model. I hope you enjoy the conversation.You can follow me on X (@ProfSchrepel) and BlueSky (@ProfSchrepel) to receive regular updates.References: Unifying complexity science and machine learning (2023) https://www.frontiersin.org/journals/complex-systems/articles/10.3389/fcpxs.2023.1235202/full The debate over understanding in AI’s large language models (2023) https://static1.squarespace.com/static/5f29a430a2b6a34680879cc0/t/672467763ec35e0639db8457/1730439030537/DK-DebateOverUnderstandingInAIsLLMs2023.pdf Darwinian demons, evolutionary complexity, and information maximization (2011) https://static1.squarespace.com/static/5f29a430a2b6a34680879cc0/t/6725792b7d0d4f0e4e7ca2fe/1730509104265/DK-DarwinianDemonsEvolutionaryComplexity%26InformationMaximization2011.pdf
--------
51:57
#15 – Larry Lessig: Code, Law, and Business Models in the Age of AI
My guest today is Larry Lessig, Professor of Law and Leadership at Harvard Law School. Larry is the author of numerous influential books and articles, including Code 2.0 (2006), which we discuss at length in this episode. If you have been listening to Scaling Theory since the very beginning, you probably remember that I cited a couple of books that changed my perception of everything in the first episode. Code 2.0 is one of these books. Larry Lessig develops what he calls the “pathetic dot theory,” in which he explains that all things are influenced by four constraints: the law, economic forces, norms, and architecture.
In this conversation, Larry and I talk about the importance of these four constraints in the digital economy and assess which ones have scaled the most in recent years. We also explore how complexity science can contribute to Larry’s theory by seeing the dots and their constraints as a complex network. We then steer our conversation toward open source in AI, examine how regulation at the hardware layer could solve software issues, and consider whether we can trust our institutions and current regulations to do so, or if we need to scale other institutions for that purpose. I hope you enjoy our discussion.
References:
Code 2.0 (2006) https://lessig.org/product/codev2/
Code (1999) https://lessig.org/product/code/
You can follow me on X (@ProfSchrepel) and BlueSky (@profschrepel) to receive regular updates.
Scaling Theory is a podcast dedicated to the power laws behind the growth of companies, technologies, legal and living systems. The host, Dr. Thibault Schrepel, has a PhD in antitrust law and looks at the regulation of digital ecosystems through the lens of complexity theory. The podcast is hosted by the Network Law Review. It features scholarly discussions with select guests and deep dives into the academic literature.
Netherlands