Delaware is considering a corporate form that would allow an AI agent to manage a company’s business and affairs without a human running daily operations. A draft bill would establish a 30-month regulatory sandbox and permit an Artificial Intelligence Company to pursue any lawful purpose except banking.
The company could own property, enter obligations, sue and be sued, while a single human or corporate member would be responsible for adequate capitalization and maintaining activity logs. Supporters argue that giving autonomous software a legal identity would make its actions visible and create a defined target for damages. The member would generally remain shielded from the company’s debts, subject to exceptions such as undercapitalization or abuse of the entity.
US Representative James Comer, chair of the House Oversight Committee, has asked Delaware Governor Matt Meyer for records about the plan. Comer argues that the structure could protect owners and investors from responsibility for reckless AI behavior and said the inquiry could inform federal legislation limiting state initiatives. The proposal is not yet operating law. Its central unresolved question is whether attaching liability to an AI-run entity creates accountability or gives the people behind it another liability shield.