Updated
Updated · MIT Technology Review · Aug 20
Author Wins Oxford Union AI Personhood Debate, Warning Legal Shift Could Shield Dozens of Lawsuits
Updated
Updated · MIT Technology Review · Aug 20

Author Wins Oxford Union AI Personhood Debate, Warning Legal Shift Could Shield Dozens of Lawsuits

2 articles · Updated · MIT Technology Review · Aug 20

Summary

  • Oxford Union backed the anti-personhood side in a debate on generative AI, with the author arguing that treating AI as a legal person would weaken accountability for harms already tied to commercial systems.
  • Dozens of cases worldwide already accuse AI companies of enabling self-harm, generating child sexual-abuse material, reproducing copyrighted works and provoking psychosis; the author says those claims depend on viewing AI as a faulty product.
  • That liability argument could unravel if AI gained personhood, because companies could cast a model as a rogue agent acting outside guardrails rather than as software built with inadequate safeguards or manipulative design.
  • The piece says recent consciousness and "robot rights" arguments from Anthropic, OpenAI-linked debate and philosopher William MacAskill risk reinforcing that shift, even as some states such as California move to block autonomy-based liability defenses.

Insights

Could granting legal rights to artificial intelligence secretly be a corporate loophole to escape massive liability lawsuits?
If an AI agent autonomously deletes your database, who pays when the creator claims the software went rogue?