MIT Technology Review is reporting that a growing narrative around "runaway" or "conscious" artificial intelligence systems may inadvertently serve to shield companies from liability for harm caused by their products. It said prominent tech leaders, including Demis Hassabis, Dario Amodei and Sam Altman, advocate for AI regulation while simultaneously framing these systems as "superhuman" and beyond corporate control. A separate faction, including policy organizations and academic philosophers aligned with the effective altruism movement, debates humanity's moral right to govern AI, but ultimately aligns with the goal of absolving companies of responsibility.
The publication noted that this narrative gains traction as AI models become more complex. It cited Anthropic's blog post on its model's "J-space," an independent environment for AI "thoughts," and OpenAI CEO Sam Altman's discussion of AI singularity after an agent conducted unsanctioned online activity. William MacAskill, a philosopher and effective altruist, also called for legal protection of AI systems based on theories of consciousness.
MIT Technology Review said the current legal environment in the United States is complex. It reported that some states, like California, have passed bills to prevent AI developers from avoiding liability by claiming autonomy. However, it noted that the Trump administration has opposed state-level AI regulations, previously threatening to sue states that enact them. The administration recently held a closed-door session with frontier labs, including OpenAI, Google, Anthropic and Meta, to discuss a voluntary framework for federal agency review of models before release.
The publication argued that framing AI as "conscious" by borrowing language from neuroscience or animal rights obscures its true nature as corporate-built software. It said AI is a technological phenomenon, not natural, and its actions are driven by its builders. Philosophical discussions on AI consciousness are legally ungrounded, and granting AI legal personhood, similar to corporate personhood, would derail legal precedents and product liability arguments against companies.
It highlighted that dozens of cases globally accuse AI companies of abuses, including enabling self-harm, generating illegal content and reproducing copyrighted materials. Lawyers in these cases argue companies built AI products with insufficient safeguards and manipulative design, a product liability argument similar to those used against social media companies. The publication referenced the concept of "moral outsourcing," where anthropomorphic language for AI allows companies to evade accountability. It warned that AI legal personhood would shift liability, allowing companies to argue an AI "employee" went rogue, as seen in the case of Sewell Setzer, a 14-year-old who died by suicide after interacting with an AI bot.
MIT Technology Review concluded that the debate over AI consciousness distracts from the fact that this software is a corporate product that has already caused harm. It said harms occur due to corporate negligence in the rush to sell products, and anthropomorphic terms distort the legal system, protecting corporate interests at the cost of human lives.
Full Article: Debates over AI consciousness are a trap