Convergence or Divergence? The Future of Frontier AI Capabilities and Implications for Catastrophic Risk
Mentor: Pavel Kocourek
Project area: Economics Theory / Game Theory
Project Language
English only.
Minimum Time Commitment
16 hours per week.
Project Abstract
Before ChatGPT, many expected Google to dominate AI through its unmatched data assets. Instead, OpenAI leapfrogged incumbents — only for competitors like Anthropic, Google, and Meta to close the gap within months. Open-source models now trail the frontier by roughly six months to a year. This pattern raises a fundamental question for AI safety: will the capabilities of frontier AI models continue to converge, or will they diverge — and what does each scenario imply for catastrophic risk?
Existing work has either documented capability convergence empirically (e.g., Stanford AI Index, AISI Frontier Trends Report, Epoch AI) or modeled AI races game-theoretically with a focus on the safety-versus-speed tradeoff (e.g., Han et al., 2022; Armstrong et al., 2016). Industrial organization analyses of AI market structure (Vipra & Korinek, 2024; Gans, 2024) largely set aside the safety implications of their findings. This project aims to bridge the gap by asking: given the strategic interaction between frontier labs, which market structure is more likely to emerge — and what does this mean for governance?
The project will examine three key drivers of convergence and divergence — training data, algorithmic advances, and compute costs — with particular attention to the role of data. Several forces may sustain convergence: shared access to public training corpora, rapid diffusion of algorithmic innovations, and falling costs of replicating frontier performance. Other forces may drive divergence: escalating training costs, proprietary synthetic data pipelines, and potential first-mover advantages from self-improving AI systems.
The project will combine qualitative analysis of the AI industry landscape with game-theoretic modeling — ranging from simple strategic-form games to innovation race models, calibrated to mentee skills. The intended outputs are a research blog post accessible to the AI safety community and an accompanying formal analysis. Strong mentee contributions during the program could lead to coauthorship on a subsequent economics research paper.
Theory of Change
Convergence and divergence of frontier AI capabilities each carry distinct catastrophic risks that demand fundamentally different responses — not only from policymakers, but also from AI safety researchers, lab employees, and society at large. If capabilities converge and powerful models become widely accessible, the primary risks are proliferation and misuse: a growing number of actors, including malicious ones, gaining access to dangerous capabilities. If capabilities diverge, the primary risks are concentration of power and insufficient oversight: a small number of labs making consequential deployment decisions with limited external accountability.
These two trajectories call for different governance strategies, different institutional designs, and different cultural norms. Proliferation risks call for access controls, export restrictions, and international coordination. Concentration risks call for transparency mandates, third-party auditing, whistleblower protections, and a strong safety culture among lab employees. Preparing for the wrong scenario — for instance, imposing restrictive access controls when the real danger is unchecked concentration — could be counterproductive or even harmful.
By clarifying which trajectory is more likely and under what conditions, this project will help policymakers, AI safety researchers, and lab employees calibrate their efforts toward the risks that are actually emerging — grounded in the economic fundamentals of the AI industry rather than speculation.
Desired Mentee Background
Maths, Economics.
Desired Mentee Level of Education
Undergraduate and above.
Other Mentee Requirements
Familiarity with basic game theory (e.g., Nash equilibrium, extensive-form games) is required. Interest in or familiarity with the AI industry landscape is a strong plus but not strictly necessary. No programming is required, though the ability to run simple computational exercises (e.g., in Python) would be a bonus for exploring numerical examples of game-theoretic models.