Stanford Computational Antitrust

Can two AI agents fix prices without ever agreeing to? Tanise Brandão and Carlos Neves of Brazi’'s competition agency, CADE, join Thibault Schrepel and Alba Ribera Martínez to walk through their paper on algorithmic collusion.
They put two large language models in a repeated Bertrand duopoly, let them price against each other for 400 rounds, and watch supra-competitive prices emerge on their own. The agents coordinate even when they cannot communicate, and some learn to cover their tracks, with one proposing that they steer clear of red flags.
The conversation turns to why the choice of model and language moves the result, and why the behavior looks less like textbook rationality than a complex dynamical system. It closes on enforcement, where collusion no longer needs a meeting of minds, and Brandão and Neves make the case for shifting liability from whether firms agreed to what they chose to expose themselves to.

https://law.stanford.edu/publications/smart-agent-based-modelling-with-llms-leveraging-large-language-models-for-a-better-understanding-of-algorithmic-collusion/

https://www.anthropic.com/research/global-workspace

In episode 39, Alba Ribera Martinez and Thibault Schrepel talk to Gustavo Augusto Freitas de Lima, Interim President of Brazil's Competition Agency (CADE).Follow the Stanford Computational Antitrust project at https://law.stanford.edu/computationalantitrust

In episode 38, Thibault Schrepel talks to Alvin Koh, Chief Executive of the Competition and Consumer Commission of Singapore (CCS).
They discuss the creation of the Data and Digital division and the operational role of instruments such as the Complaint Analytics Tool and the AI Verify Toolkit in investigative prioritization and compliance assessment. They also explore the application of computational methods in major enforcement initiatives, including the Grab–Uber merger and the Price Transparency Guidelines, before considering the implications of these practices for the future of competition enforcement in Singapore and the ASEAN region.
Follow the Stanford Computational Antitrust project at ⁠https://law.stanford.edu/computationalantitrust

In episode 37, Teodora Groza and Thibault Schrepel talk to Nuno Cunha Rodrigues, President of the Portuguese Competition Authority since 2023. They discuss the tools developed by the agency, how its lawyers work with economists and computer scientists, and the future of the field. Follow the Stanford Computational Antitrust project at https://law.stanford.edu/computationalantitrust

In episode 36, Thibault Schrepel talks to Nancy Scola, longtime technology and politics journalist & reporter.
Thibault and Nancy discuss stories that deserve more attention from computational antitrust researchers, how institutions adapt (or fail to adapt) to technological change, and how journalism and computational tools can complement one another. They also consider whether “computational antitrust journalist” could become a job description in the future, and other related topics.
Follow the Stanford Computational Antitrust project by subscribing to our newsletter at ⁠https://law.stanford.edu/computationalantitrust