How should science work in the age of AI?

Primary supervisor

Julian Garcia Gallego

Research area

Agents and Decision-Making

Science depends on institutions. Journals decide what is published, reviewers assess work produced by their peers, funders choose which questions receive support, and universities reward some kinds of contribution more than others. These arrangements shape what scientists study and how carefully they study it.

AI changes this system because it can assist with writing, analysis, experimentation and review. Producing a plausible paper may become much cheaper while judging whether a result is important and reliable remains difficult. How should scientific institutions respond? Which uses of AI should be disclosed? Can peer review cope with a large increase in submissions? How should credit be assigned when research is partly automated? Could new publishing or funding mechanisms produce better science?

This project uses game theory and computational models to study science as a social system. The precise question will be developed with the student and could focus on publishing, peer review, research funding, scientific careers, collaboration, or the norms governing AI use. Projects may also use data from publications, citations or scientific careers to test the implications of a model.

References

  • Fortunato, S. et al. “Science of science.” Science 359 (2018): eaao0185. https://doi.org/10.1126/science.aao0185
  • Zollman, K. J. S., García, J. and Handfield, T. “Academic journals, incentives, and the quality of peer review: a model.” Philosophy of Science 91(1) (2024): 186–203. https://doi.org/10.1017/psa.2023.81
  • Aczel, B. et al. “The present and future of peer review: Ideas, interventions, and evidence.” PNAS 122(5) (2025): e2401232121. https://doi.org/10.1073/pnas.2401232121

Required knowledge

  • An excellent academic record in computer science, economics or a cognate field. An Honours degree with H1 or equivalent is required for Graduate Research admission.
  • An interest in game theory and in how scientific institutions work.
  • Solid mathematical modelling and good programming skills.
  • Excellent written and verbal communication skills.

Learn more about minimum entry requirements.