From the Dark Side to the Bright Side: Exploring Algorithmic Antitrust Compliance

Abstract

If pricing algorithms could autonomously collude, can they be made automatic antitrust compliant as well? That is the question many have started pondering after a series of public comments by EU competition officials in recent years. In this paper, I propose a multi-faceted approach to algorithmic compliance. I draw lessons from the recent AI literature and discuss some potential technical frameworks for compliant algorithmic design. [Note: this paper is being superceeded by my chapter on algorithmic collusion and compliance for the Report on Digital Economy to be published by the Global Antitrust Institute at George Mason law.]