Humanoid robots are coming. Insurance isn’t ready, yet
Life-like machines are moving into homes, factories and hospitals. They bring risks no policy was written for.
InsureStory
Some technologies upgrade how we live. Others rewrite the rules. Humanoid robots are the second kind. As they move from labs into homes, factories, hospitals and public spaces, insurance faces risks that don’t behave like human risks, liabilities that don’t fit existing law, and policies never built for machines that make their own decisions.
The opportunity is huge. Estimates put the economic impact in the trillions, with robots able to perform more than half of today’s work tasks, from logistics and elderly care to hospitality and manufacturing.
Humanoids don’t get tired or distracted. But they do malfunction, misread instructions, make autonomous decisions, suffer software glitches and get hacked.
That creates four new exposures. Robot accident liability: if a humanoid drops a pallet or injures a customer, who pays: the employer, the manufacturer, the programmer, the owner, or the robot’s own policy? This is the dawn of robot torts. Algorithmic liability: when an autonomous decision causes a loss, negligence and human error no longer explain it. Cyber and sensor failure: a hacked humanoid isn’t just a data breach, it’s a physical safety threat. And operational reliability: robots need constant calibration and updates, so cover must account for predictive maintenance, not just accidents.
Then comes the legal question. Law has always assumed that humans cause harm and humans bear responsibility. Can a robot be a legal actor? Who is at fault when a machine decides on its own? Regulators have barely started to answer.
Insurers that start now, building new wordings, new data partnerships with manufacturers and new ways to price machine behaviour, will shape this market. The rest will be catching up to it.