An AutoModerator is a bot designed to automate various moderation tasks that require little or no human judgement. It can watch the new/spam/comments/report queues of any subreddit it moderates and take actions on submissions and comments based on defined conditions. This includes approving or removing them . . . . It is effectively fairly similar to reddit’s built-in spam-filter, but [also] allows for conditions to be defined specifically instead of just giving vague hints by removing/approving. Its decisions can always be overridden by human mod[erators], exactly like an existing filter.
Assistant Professor of Law, Drexel University Thomas R. Kline School of Law; Affiliated Fellow, Yale Law School Information Society Project. I am very grateful to RonNell Andersen Jones, Rebecca Crootof, James Grimmelmann, Anil Kalhan, Jonathan Marks, Christopher Reed, Daniel Susser, and Ari Ezra Waldman for their helpful comments and feedback on this project. Portions of this Article were presented at the Cornell International Law Journal’s Symposium, the Penn State Law School Works-in-Progress Workshop, and Yale Law School’s Information Society Project. My thanks to the student organizers of the symposium and the editors of the Cornell International Law Journal. This Article reflects developments through December 2019, when it was finalized for publication. All errors are my own.