Shareholder Democracy with AI Representatives

Proposes AI-enabled representatives trained on individual shareholder preferences to vote on their behalf in corporate governance. Addresses the problem that mutual funds concentrate voting power among few asset managers who lack insight into individual preferences. Argues this approach could outperform both pass-through voting (which has low participation) and “investor assemblies” (randomly selected shareholders). Suggests shareholder democracy offers a compelling test bed for AI-enabled representation, with potential to predict not just current preferences but how investors might vote with more time and resources.