Build an economic theory of trade secret and confidentiality law for AI

Develop a robust economic account of trade secret and confidentiality law that complements existing analyses of patent and copyright, articulating incentives, costs, and welfare tradeoffs relevant to AI foundation model development, secrecy, and regulatory access.

Background

The chapter notes that while economics provides careful analyses of patent and copyright, there is no correspondingly robust account for trade secrets and confidentiality—critical for AI given proprietary model training and evaluation constraints. A rigorous economic framework is needed to inform policy on balancing innovation incentives against societal needs for transparency and oversight.