Continuing to Build a Global, Multidisciplinary Community of Knowledge
The 2026 cohort of the AI Economy Institute (AIEI) brings together researchers from across North America, Europe, the Middle East, Africa, and Asia, spanning:
- Economics and economic geography
- Organizational behavior and labor economics
- Computer science and AI systems
- Public policy and governance
This interdisciplinary composition reflects AIEI’s core premise: understanding the AI economy requires examining how AI capabilities move through institutions, reshape work and production, and inform policy.
Cohort 3 marks a turning point in AIEI’s research trajectory. Cohort 1 examined how higher education can prepare people for an AI-enabled economy. Cohort 2 expanded the lens to education systems and institutional pathways for diffusion. Cohort 3 moves our point of inquiry to the frontier, asking how AI is reshaping firms, labor markets, and economic structure in real time.
From Questions to Framework: How Cohort 3 Is Structured
Cohort 3 begins with a set of five priority research themes that define the initial question.
From there, Fellows submit proposals that interpret and engage with these questions. While Fellows initially align to one or more themes, AIEI then reorganizes the research portfolio based on the insights that emerge.
In this way, Cohort 3 is both guided and adaptive:
- The five themes establish the research agenda
- The Fellows’ work reshapes that agenda into a more empirically grounded synthesis
Priority Research Themes
The cohort is anchored in five interlocking themes:
- Productivity at the Frontier and Firm-Level Transformation. How AI is reshaping production and organizational design
- Occupational Change, Leadership Expectations, and Workforce Transformation. Evolving skills, roles, and leadership in an AI-enabled workforce
- Economic Geography and Diffusion. Regional and market-level spillovers
- When History Rhymes. GPT analogues and structural shifts
- Forecasting the Frontier. Capabilities, diffusion, and labor market signals
Together, these themes explore the idea that AI’s economic impact will be determined not simply by technological advances, but by the deployment of capabilities within firms, across labor markets, and throughout global economic systems.
Beyond Adoption: Studying the AI Economy at the Frontier
The central question for Cohort 3 is: if AI capability is advancing rapidly, what determines who adopts, how quickly, and with what consequences for firms, workers, and economic structure?
This cohort moves beyond the question of whether AI is being adopted to examine how the AI economy is becoming visible inside production systems, labor markets, and global infrastructure.
A Shift in Focus: From Diffusion to the Frontier
Cohort 3 represents a deliberate shift in emphasis:
- From education systems → firms as the primary site of transformation
- From access to AI → constraints on scaling and deployment
- From potential impacts → observable organizational and labor market effects
The operating premise is that AI diffusion is no longer hypothetical. It is fast, uneven, and empirically observable, particularly within frontier firms and early-adopting economies.
From Themes to Synthesis: Four Emerging Areas of Inquiry
As Fellows engaged with the five themes, their proposals surfaced common questions and shared lines of inquiry, reshaping the research into four empirically grounded streams of work.
1. The Adoption Frontier: What Governs the Speed of AI Uptake
AI adoption is not determined by capability alone. It depends on the conditions that make deployment feasible: infrastructure, compute, energy, and the structure of inference markets.
Even as models improve rapidly, adoption is constrained by economic and operational realities. The pace of diffusion reflects feasibility as much as progress.
Contributing Senior Fellows include:
- Edoardo Maria Acabbi (University of Mannheim, Germany) and Luca Mazzone (University of Montreal, Canada) on supply constraints and the “feasibility frontier”
- Caspar David Peter (Erasmus University, Netherlands) on historical analogues to technological disruption
- Ilan Strauss (The AI Disclosure Project) on AI diffusion constraints, including affordability, demand, wage share, financing capacity, and capital intensity
- Daniel Yue (Georgia Institute of Technology) and Frank Nagle (MIT) on competition and lock-in in inference markets
2. Uneven Diffusion: Why AI Does Not Spread Equally
AI does not diffuse evenly across firms, regions, or economies. Differences in capital, talent, institutions, and policy environments produce persistent gaps between frontier and lagging adopters.
Diffusion is also shaped by less visible factors such as informal economies, cross-border networks, and firm-level relationships, which influence where and how AI takes hold.
Contributing Senior Fellows include:
- Pierre-Alexandre Balland and Laura Nurski (CEPS, Belgium)) on a “multi-AI economy” across cities
- Nuriye Melisa Bilgin (Koc University, Turkey) and Gianmarco Ottaviano (Bocconi University), on governance and legal uncertainty as adoption barriers in emerging economies
- Wesley Rosslyn-Smith (University of Pretoria, New Zealand) on informal economies
- Johannes Wachs (Cornivus University, Hungary) on network connectivity
- Nataliya Wright (Columbia University) on firm-level scaling and global diffusion
3. Inside the Frontier Firm: How Organizations Actually Change
A central question is whether individual productivity gains from AI scale into measurable firm-level outcomes.
This work examines how AI reshapes teams and workflows, role design and task allocation, and the balance between human and machine capabilities. It also investigates whether AI introduces new coordination challenges alongside productivity gains.
Contributing Senior Fellows include:
- Salman Khan and Mustafa Afcan (MBZUAI, UAE) on occupational restructuring (in collaboration with the World Bank)
- Friederike Mengel (University of Essex, United Kingdom) and Christoph Siemroth (University of Exeter, United Kingdom) on team productivity and collaboration
- Yingfei Wang (University of Washington) on learning-aware delegation
4. Agentic Labor Markets: When AI Becomes an Actor
AI systems are increasingly moving from tools to participants in economic activity. As they begin to hire, manage, and evaluate workers, they reshape core labor market functions.
This raises new questions about firm decision-making, labor market dynamics, and governance and accountability. It also introduces risks related to transparency, bias, and worker agency.
Contributing Senior Fellows include:
- Serena Booth (Brown University) on worker voice and governance
- Meeyoung Cha (KAIST, South Korea and Max Plank Institute) on AI agents as employers
- Brian Jabarian (Carnegie Mellon University), Luca Henkel (Erasmus University, Netherlands), and Pellumb Reshidi (Florida State University) on agentic hiring systems
Interpreting the Frontier: Implications for Policy and Strategy
Taken together, the research suggests that the AI economy will not follow a single trajectory. Instead, it will be defined by a set of persistent tensions:
- The gap between capability and deployability
- Productivity gains alongside organizational fragility
- Competition between open access and market concentration
- Trade-offs between automation and human skill development
- Uneven diffusion across firms and regions
Looking Ahead
By integrating the initial research themes with insights emerging from the Fellows’ work, Cohort 3 provides a clearer, evolving picture of how AI is reshaping the economy.
The resulting body of research will inform policymakers, firms, and institutions seeking to make decisions in a rapidly changing technological and economic landscape, advancing AIEI’s mission to build a rigorous, global understanding of the AI economy.
Artificial intelligence 
Economics