This article employs a machine learning classifier, trained on a curated dataset comprising blog posts and other antitrust-related texts, to generate an index that categorizes texts along an ideological spectrum that ranges from Chicago School thinking at one end to a Neo-Brandeisian perspective at the other. The index is used to explore a diverse corpora of antitrust texts, shedding light on the evolving landscape of beliefs and attitudes over time. These corpora include antitrust blogs, an official Federal Trade Commission (FTC) blog, speeches by FTC commissioners and Department of Justice (DOJ) officials, as well as antitrust guidelines and legislation. The results show the arc of antitrust in the US, that is, the shift from interventionist Brandeisian thinking until the late seventies, followed by several decades of Chicago School-inspired, more permissive antitrust, and more recently, a discernible reversal towards a more (Neo-) Brandeisian position.