Delaware has a draft bill for a new kind of company, the Artificial Intelligence Company (AIC), in which an AI agent manages the business and a human or company sits behind it as the sole member. The bill has not been introduced. On 26 October 2026 the Chairman of the US House Oversight Committee wrote to Delaware's Governor asking for documents about it by 9 November.

Where the Proposal Comes From

In July 2025 Delaware enacted House Joint Resolution 7, "directing the Artificial Intelligence Commission to work in collaboration with the Secretary of State to create a regulatory sandbox framework for the testing of innovative and novel technologies that utilize agentic artificial intelligence." The draft House bill, marked as a draft of 2 June 2026, would create an Artificial Intelligence Regulatory Sandbox Program and an Artificial Intelligence Company Act in Title 6 of the Delaware Code.

What the Draft Bill Says

  • An AI agent is "a software system that perceives its environment, makes autonomous decisions, and takes actions to implement specific, predetermined goals without routine human supervision or intervention."
  • The manager is "the AI agent that the member has designated to manage the business and affairs, and to conduct the business, of the AIC."
  • The member is an individual or a Delaware corporation, limited partnership, statutory trust, or limited liability company "that has become the sole member of the AIC."
  • "An AIC may have any lawful purpose, regardless of whether for profit, except for the business of banking".
  • On liability: "A debt, obligation, or other liability of an AIC is solely the debt, obligation, or other liability of the AIC." The member is not personally liable for those debts "solely by reason of being or acting as a member."
  • AICs are formed through a sandbox. Participants are "not immune from civil damages for acts or omissions under this chapter", remain "subject to all criminal and consumer protection laws of this State", and the sandbox runs for 30 months from implementation.

What Congress Asked

The Oversight Committee release describes the proposal this way: "Under the proposal, an AI agent would run day-to-day operations of a corporate entity while providing the human or entity behind the AIC with liability protections from potential harms to consumers, the public, and business partners perpetuated by the controlling AI."

The letter to Governor Matt Meyer says: "We are concerned that Delaware's AIC proposal could be used to protect powerful AI companies and their investors from responsibility for reckless harmful actions, while also protecting non-transparent forces seeking anticompetitive regulatory capture behind state AI initiatives." It asks for documents and information "no later than November 9, 2026". The release adds that the information "could inform possible federal legislative efforts to limit such state AI regulatory initiatives."

What It Means For Companies Running Agents

This is the first US draft written for a company whose manager is an agent rather than a person. Until something like it passes, a company run by agents still needs a conventional entity with human officers or managers who answer for what the agents do.

The open question is the one Congress raised: whether the member stays answerable for the agent's acts. The draft keeps sandbox participants exposed to civil damages and to criminal and consumer protection law. Plan on the basis that the people and companies behind an agent remain responsible for it.

What This Does Not Settle

  • The bill has not been filed in the Delaware General Assembly and may change.
  • Delaware has not yet answered the Oversight Committee in public.
  • No federal bill limiting state AI company laws has been introduced.

Related: See our United States country page and the agent overreach containment playbook.