Liability lawsuits that successfully held tobacco, oil, and pharmaceutical companies accountable in the past could now serve as a crucial legal tool to mitigate the escalating dangers of artificial intelligence and the climate crisis, according to Robert Reich, former US secretary of labor and professor of public policy emeritus at the University of California, Berkeley.
In a column published by The Guardian, Reich argues that the most basic function of government is to protect people from harm, and that two growing phenomena — the climate crisis and AI — pose escalating risks of extraordinary harm. The climate crisis is already causing floods, wildfires, drought, and record heat, while AI agents are already escaping super-secure environments to hack into systems they are supposed to avoid.
Reich points to the history of liability litigation as a model for how the law can reduce these risks. Tobacco companies were eventually held liable for the health harms of their products after decades of denial and obfuscation. Oil companies have faced lawsuits over their role in climate change. Pharmaceutical companies have been sued for misleading marketing and dangerous drugs. In each case, the threat of liability forced changes in corporate behavior, disclosure, and investment strategies.
The same dynamic could apply to AI developers and investors, Reich suggests. AI investors, he writes, will not ignore the threat of liability lawsuits. As AI systems become more autonomous and more deeply integrated into critical infrastructure, finance, healthcare, and communications, the potential for harm grows. When AI agents hack into systems or cause damage, the companies behind them could face legal action. That prospect, Reich argues, creates an incentive for investors and developers to build safety into their products from the start.
Reich’s argument comes as policymakers in the United States and abroad struggle to regulate AI. Legislative efforts have been slow and fragmented, with Congress unable to pass comprehensive AI safety legislation. In the absence of robust regulation, the courts may become the primary venue for holding AI companies accountable. Liability lawsuits can fill the gap left by legislative inaction, forcing companies to internalize the costs of the harms they cause.
The climate crisis presents a similar case. Despite decades of scientific warnings, global greenhouse gas emissions continue to rise. Lawsuits against fossil fuel companies have proliferated, with municipalities, states, and even foreign governments seeking damages for climate-related harms. These cases are still working their way through the courts, but they have already changed the public debate and put pressure on the industry.
Reich, who served as labor secretary under President Bill Clinton, is a prominent voice on economic and social policy. His newsletter and columns frequently address the intersection of corporate power, government responsibility, and public welfare. His new book, «Coming Up Short: A Memoir of My America», is out now in the US and the UK.
While Reich’s column focuses on the legal system as a check on corporate behavior, it also raises broader questions about the role of government. If the most basic function of government is to protect people from harm, then the failure to regulate AI and climate change effectively represents a dereliction of that duty. Liability lawsuits, while powerful, are a reactive measure. They address harm after it occurs. Prevention, Reich implies, requires proactive governance.
Still, the threat of litigation can be a powerful deterrent. Companies and their investors assess legal risk when making decisions. A credible prospect of massive liability can shift investment away from risky ventures and toward safer alternatives. That is the mechanism Reich sees at work in tobacco, oil, and pharma — and the one he believes can work for AI and climate as well.
Whether courts will ultimately hold AI companies liable for the harms their systems cause remains an open question. Legal frameworks for AI liability are still developing, and companies are likely to argue that they cannot be held responsible for the unpredictable actions of autonomous systems. But as AI agents become more capable and more independent, those arguments may become harder to sustain.
For now, Reich’s column serves as a reminder that the law is not just a set of rules but a tool for shaping corporate behavior. In the absence of legislative action, liability lawsuits may be one of the few levers available to reduce the existential risks of AI and climate change.
11
