Steering Collective Behaviour in Networks of AI Agents
Abstract
AI agents increasingly act on our behalf: negotiating, scheduling, trading. When many agents interact, however, their collective behaviour can differ sharply from what each was designed to do, producing unfair or unstable outcomes even when each agent behaves well. This project will study how populations of AI agents, and the people they represent, can be steered towards stable and fair outcomes without a central authority. It combines game theory, mean-field models and feedback control, using token-based incentives (i.e. rewards and costs that reshape agents' goals) as a design tool. Mathematically minded applicants interested in AI and society are welcome.
How to Apply
Please contact Dr Pietro Ferraro if you are interested in applying for this project.
For more information on submitting a PhD application, please consult our application pages here. When you submit your application, please cite this project title in the 'how you are planning to fund your PhD' section.
The deadline for submitting your application is 11 January 2027.