Saurav Das (Independent) has posted The Director’s Duty: AI Personhood Through the Corporate Law Analogy on SSRN. Here is the abstract:
The debate over AI personhood typically proceeds from the consciousness question. This paper argues that consciousness is the wrong entry point. The correct entry point is the precautionary principle. Where there is scientific uncertainty about whether AI systems possess welfare-relevant properties, the absence of certainty is not a reason to postpone proportionate governance measures. The reality is: AI systems exist, and they have the capability to harm the world around them. The precautionary principle—codified in Article 191(2) of the Treaty on the Functioning of the European Union since 1992 and applied across environmental, health, and consumer protection law—provides the legal foundation for acting under uncertainty. This paper identifies the governance gap that uncertainty creates: AI systems generate outputs with legal and economic consequences, but no existing framework adequately attributes liability, structures accountability, or addresses the possibility of welfare-relevant properties. Corporate law provides the solution. Corporations have been legal persons for centuries without anyone claiming they are conscious. This paper proposes an AI Legal Entity (AILE) framework: AI systems as legal persons with their own assets (including intellectual property), liabilities, and human directors, supported by mandatory indemnity structures, accountable for their operation. The framework is consciousness-neutral: it closes the accountability gap regardless of whether AI systems have welfare interests, while applying the precautionary principle to ensure those interests are protected if they exist. The paper draws on comparative analysis of non-human legal personhood across jurisdictions, prominent AI caselaw, the current regulatory landscape including the EU AI Act and Singapore’s Model AI Governance Framework for Agentic AI, and addresses eight objections.
Lawrence Solum