•  
  •  
 

Publication Date

Fall 6-25-2026

Abstract

Artificial intelligence is destabilizing one of tort law’s most influential organizing principles: the Cheapest Cost Avoider (“CCA”). Classical CCA analysis emerged in relatively bounded accident settings, where the actor best positioned to prevent harm was often also best positioned to evaluate the relevant cost–benefit tradeoffs. Algorithmic systems disrupt this alignment. In AI-driven environments, harms increasingly emerge from layered socio-technical ecosystems involving developers, platform providers, institutional deployers, and opaque model architectures. Under these conditions, the most visible human actor at the point of injury—the physician, driver, or loan officer—often appears to be the CCA, yet lacks meaningful control over the system-level risks that precipitated the harm.

This Article argues that tort law must move beyond the CCA paradigm and reorient liability around a new organizing concept: the Best Algorithmic Risk Governor (“BARG”). Building on Calabresi and Hirschoff’s best decision maker (“BDM”) framework and the economic logic of the Hand formula, the Article contends that liability should concentrate on the actor best positioned to observe, evaluate, and govern systemic algorithmic risk at scale. The Article develops functional markers for identifying BARGs, including control over training data, model architecture, monitoring infrastructure, and population-level risk observability.

Using medical AI, autonomous vehicles, and algorithmic credit systems as case studies, the Article demonstrates how contemporary tort doctrine systematically misallocates responsibility by focusing on downstream human discretion rather than upstream governance power. It concludes that tort law’s efficacy in the algorithmic age lies not in assigning blame for isolated accidents, but in structuring incentives for continuous institutional risk governance.

Download

Share

COinS