Without clear government standards for AI safety, there is no "safe harbor" from lawsuits. This makes it likely courts will apply strict liability, where a company is at fault even if not negligent. This legal uncertainty makes risk unquantifiable for insurers, forcing them to exit the market.
Related Insights
The insurance industry acts as a powerful de facto regulator. As major insurers seek to exclude AI-related liabilities from policies, they could dramatically slow AI deployment because businesses will be unwilling to shoulder the unmitigated financial risk themselves.
Existing policies like cyber insurance don't explicitly mention AI, making coverage for AI-related harms unclear. This ambiguity means insurers carry unpriced risk, while companies lack certainty. This situation will likely force the creation of dedicated AI insurance products, much as cyber insurance emerged in the 2000s.
AI companies engage in "safety revisionism," shifting the definition from preventing tangible harm to abstract concepts like "alignment" or future "existential risks." This tactic allows their inherently inaccurate models to bypass the traditional, rigorous safety standards required for defense and other critical systems.
Drawing from the nuclear energy insurance model, the private market cannot effectively insure against massive AI tail risks. A better model involves the government capping liability (e.g., above $15B), creating a backstop that allows a private insurance market to flourish and provide crucial governance for more common risks.
Insurers like AIG are seeking to exclude liabilities from AI use, such as deepfake scams or chatbot errors, from standard corporate policies. This forces businesses to either purchase expensive, capped add-ons or assume a significant new category of uninsurable risk.
An FDA-style regulatory model would force AI companies to make a quantitative safety case for their models before deployment. This shifts the burden of proof from regulators to creators, creating powerful financial incentives for labs to invest heavily in safety research, much like pharmaceutical companies invest in clinical trials.
Insurers can price a single large loss. What they cannot price is a single AI model, deployed by thousands of customers, having a flaw that leads to thousands of simultaneous claims. This "systemic, correlated" risk could bankrupt an insurer.
AI and big data give insurers increasingly precise information on individual risk. As they approach perfect prediction, the concept of insurance as risk-pooling breaks down. If an insurer knows your house will burn down and charges an equivalent premium, you're no longer insured; you're just pre-paying for a disaster.
The approach to AI safety isn't new; it mirrors historical solutions for managing technological risk. Just as Benjamin Franklin's 18th-century fire insurance company created building codes and inspections to reduce fires, a modern AI insurance market can drive the creation and adoption of safety standards and audits for AI agents.
Technological advancement, particularly in AI, moves faster than legal and social frameworks can adapt. This creates 'lawless spaces,' akin to the Wild West, where powerful new capabilities exist without clear rules or recourse for those negatively affected. This leaves individuals vulnerable to algorithmic decisions about jobs, loans, and more.
