Designing Loyalty: AI Agents and Conflicts of Interest
Agency law and fiduciary theory have circled this point for a while
Stanford HAI has an issue brief out on AI agents and conflicts of interest: "Designing Loyalty."
The argument: developers and deployers of consumer AI agents should be treated as fiduciaries who owe users a duty of loyalty, especially in high-stakes areas like health and finance, modeled on ERISA's domain-limited version. The duty can't live in the software, so it lands on the deploying company, and disclosure alone won't save it.
Notice the convergence. Agency law and fiduciary theory have circled this point for a while, and now the privacy world lands on it too. Three fields arriving at the same doorstep.
Research: Stanford HAI, Policy & Society Issue Brief, August 2026
Designing Loyalty: AI Agents and Conflicts of Interest | Stanford HAI
This brief examines the conflict-of-interest risks posed by AI agents and calls for imposing a duty of loyalty on developers and deployers.
Ella Genasci Smith