
What happened
Delaware is proposing a new legal entity form called the AIC (artificial intelligence company), designed to allow autonomous AI agents to operate businesses with no human at the helm. The state plans to test AICs within a regulatory sandbox overseen by a committee that includes the Delaware Secretary of State, Attorney General, Chief Justice, and the chair of the state's AI Commission, with the program set to sunset in 30 months.
Why it matters
As AI agents begin to contract, pay suppliers, and transact independently, wrapping them in a recognizable legal form creates accountability and makes their conduct visible and traceable. A Delaware AIC has a single human member responsible for capitalization, must keep activity logs, and can sue and be sued in its own name—establishing a clear target for responsibility and damages if something goes wrong. Without a legal framework in the US, the activity may migrate offshore to anonymous infrastructure beyond any court's reach.
What to watch
AICs must meet capitalization requirements, disclose to counterparties that they are authorized test entities, and notify parties when the test will end and how to file complaints. Banking is excluded from the sandbox. Officials can suspend or revoke AIC authorization, and the state can ask the Court of Chancery to dissolve an AIC. Consumer protection and criminal law apply in full.
Summaries like this, in your inbox every morning.
Delaware's proposal addresses a question that has grown from theoretical to practical: as AI agents become capable of independent contracting and commerce, what legal framework should govern them? The state's 2023 Science essay observed that nothing in existing state law clearly prevents an autonomous agent from operating a company, and asked what should be done as the technology matured. Three years later, with AI capabilities substantially advanced, Delaware is offering an answer: create a recognizable legal form that makes autonomous actors legible to law.
The AIC framework is designed to pilot autonomous commerce under controlled conditions. By requiring a single human member, capitalization standards, activity logs, and disclosure to counterparties, the structure creates accountability while insulating the human member from liability (except in cases of undercapitalization or fraud). The regulatory sandbox—with explicit sunset, oversight by senior state officials, and the ability to revoke authorization or dissolve an entity—ensures that autonomous systems operate "in daylight, under supervision, with capital tied to liability." This approach mirrors historical precedents: the modern corporation, the series LLC, and the public benefit corporation were each contested and then widely adopted.
The stakes for the US legal system are framed as significant. If American law does not provide an accountable pathway, the body suggests autonomous commerce will "migrate offshore and onto anonymous infrastructure beyond the reach of any court." Delaware's position as "the state with the deepest experience in entity formation and governance" positions it to establish norms for governing AI agents within the American legal tradition, where oversight is possible. The framework is being developed as a public-private partnership led by Norm Ai.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Cammarata reinterprets InceptionV1's cat-car neuron as a V-shape detector, not two unrelated concepts

Google Deepmind's Gemini 4 Argon scored 97.5 percent on ARC-AGI-3 at a cost of $4,800, second only to GPT-6 As…

Microsoft's Surface Laptop Ultra, using Nvidia's RTX Spark platform, starts at $2,599 (up to $5,899)

Meta's new commercial terms, effective October 30th, appear to let it crawl and train AI on content from websi…

Vercel data published Thursday shows TypeSafe AI's Jev, a lightweight decision model for choosing among predet…

Reflection AI introduced Beam, an open-weight model for agentic coding and long-horizon tasks, with 501 billio…
